Ante Aikio & Jussi Ylikoski Sámi University of Applied ... · languages and in Hungarian. Even though such case systems are characteristic of many modern Uralic languages, they are

Ante Aikio amp Jussi Ylikoski Saacutemi University of Applied Sciences amp

UiT The Arctic University of Norway

Fenno-Ugrica Suecana Nova Series bull 15 (2016) bull 59-158 bull copy Aikio A amp Ylikoski J 2016

The origin of the Finnic l-cases1

Ante Aikio amp Jussi Ylikoski

Abstract The Finnic languages among them Finnish and Estonian are well known for their

large inventories of cases As large case systems tend to develop especially through

agglutination of adpositions it is noteworthy that none of the thirteen cases reconstructed for

Proto-Finnic have traditionally been considered to derive from adpositional phrases

However in this paper such an explanation is presented for the origin of the Proto-Finnic

external local cases or the so-called l-cases ie the adessive (-llA lt -l-nA) the ablative (-l-

tA) and the allative (-l-en) The element -l- has traditionally been equated with a

derivational suffix indicating locality but against the received view this paper argues that the

endings emerged via agglutination of the Proto-Uralic postpositions uumll-nauml [on-LOCATIVE]

uumll-tauml [on-ABLATIVE] and uumlli-ŋ [on-LATIVE] based on the relational noun root uumll(i)-

lsquolocation onaboversquo The argumentation is based on rich comparative data from the Saami

Mordvin Permic and Samoyed branches of the Uralic language family Through a thorough

analysis of phonological morphological syntactic and semantic properties of the Finnic l-

cases and their proposed cognates it is argued that the received view on the origin of the l-

cases must be rejected as an illegitimately canonized hypothesis that was never tested through

systematic application of the comparative method Instead the comparative analysis strongly

supports the new hypothesis of the postpositional origin of the l-cases

Keywords Finnic languages local cases grammaticalization Saami languages Uralic

languages

1 This paper is an expanded English version of a paper originally published in North Saami (Aikio amp Ylikoski

2007) and ultimately based on a presentation at the meeting of the Finno-Ugrian Society in Helsinki on January

20th 2006 We wish to thank those present at the meeting for their questions and remarks as well as the

anonymous reviewer of Fenno-Ugrica Suecana and a number of colleagues for valuable comments on various

versions of this paper over the years


60

1 Introduction

2 A review of previous research

3 The origin of Finnic l-cases in light of the comparative method

31 A new functionally based hypothesis

32 l-cases compated against Saami al-postpositions

321 A qualitative look at the material

322 Quantitative analysis

33 Comparing l-cases to their Permic and Samoyed equivalents

34 On the phonological and morphological development of the l-cases

35 Additional evidence from Finnic

36 Parallels from other languages

37 Comparing the previous lA-theory and the new uumll-theory

4 On the secondary functions of the l-cases

41 The instrumental use of the adessive

42 The possessive use of the l-cases

5 What is left of the lA-theory

6 Discussion and conclusion

1 Introduction

In linguistic literature the Uralic languages are well-known for their large case inventories

Extensive case systems consisting of over ten cases are found in Finnic Mordvin and Permic

languages and in Hungarian Even though such case systems are characteristic of many

modern Uralic languages they are not considered primary to the language family only six

cases are traditionally reconstructed to Proto-Uralic (Janhunen 1982 30ndash31) which is not a

typologically unusual number Hence the question of how the extensive case systems

characteristic of many branches of the family have developed has become a central research

problem in Uralic historical morphology

In the western part of the language family the case system evidently became enriched

already at an early period Through a comparison of Saami Finnic and Mordvin languages

one can reconstruct as many as thirteen cases or case-like suffixes which are reflected in at

least two of these three language branches (see Table 1) The most important innovation

common to these languages (and to Mari as well) involves a reorganization of the local case

______________________________________________Origin of Finnish l-cases

61

system It is assumed that Uralic originally had a tripartite system of local cases a static

locative case (-nA) an ablative case signifying movement away from a point of reference (-

tA) and a directional ldquolativerdquo case signifying movement to a point of reference (-ŋ) In the

western branches of Uralic (Saami Finnic Mordvin Mari) these cases are attested in

predominantly grammatical functions and the lative has largely lost its productivity it is

preserved as a productive case only in Mordvin The local functions were apparently taken

over by a new set of local cases built with a so-called coaffix -s- inessive -s-nA elative -

s-tA and illative -s or -s-in (perhaps from earlier -s-iŋ in Mordvin languages the illative

ending is merely -s) As recently argued by Ylikoski (2016) it is likely that the western

Uralic coaffix -s- ultimately goes back to Proto-Uralic and is cognate with the Samoyed

local case coaffix -ntə- In any case the development of these so-called s-cases evidently

antedates the topic of the present paper the emergence of the so-called l-cases in Proto-

Finnic

Case Suffix Saami

languages

Finnic

languages

Mordvin

languages

nominative -Oslash (pl -t) + + +

genitive -n + + +

accusative -m + + +

essive -nA + + ndash

translative -ksi (+) + +

partitiveablative -tA + + +

lative -ŋ ( ~ -k -n) (+) (+) +

prolative -ko (+) ndash +

inessive -snA + + +

elative -stA + + +

illative -s ~ -sin + + +

comitative -jnV + + +

abessive -ptAk + + ndash

Table 1 Reconstructed case endings in Saami Finnic and Mordvin languages The symbol (+) indicates that the

ending is found only in adverbs or relic forms but not as a productive part of the case system

There is also a crucial feature which distinguishes the local case systems of most Finnic

languages from those of Saami and Mordvin (and almost all other Uralic languages) an

opposition between the so-called lsquointernalrsquo and lsquoexternalrsquo local cases In addition to the

lsquointernalrsquo local cases formed with the coaffix -s- a series of lsquoexternalrsquo local cases that are


62

formed with the coaffix -l- emerged in Proto-Finnic In contrast to this traditional

terminology we prefer to call these s-cases and l-cases according to the coaffix in each

series2 The paradigm of local case endings reconstructed for Proto-Finnic can be seen in

Table 2

CASE PROTO-FINNIC PRE-FINNIC

S-CASES LOCATIVE inessive -ssA lt -s-nA

SEPARATIVE elative -stA lt -s-tA

DIRECTIONAL illative -hVn lt -s-in

L-CASES LOCATIVE adessive -llA lt -l-nA

SEPARATIVE ablative -ltA lt -l-tA

DIRECTIONAL allative -l(l)en lt -l(l)-in

Table 2 The Proto-Finnic local case endings

For the sake of readers unacquainted with the case systems of Finnic languages the semantic

opposition between the s-cases and the l-cases can be illustrated with the following set of

Finnish examples (see Table 3)

vuode lsquobedrsquo talo lsquohousersquo

S-CASES INESSIVE vuoteessa lsquoin the bedrsquo talossa lsquoin the housersquo

ELATIVE vuoteesta lsquoout of the bedrsquo talosta lsquoout of the housersquo

ILLATIVE vuoteeseen lsquointo the bedrsquo taloon lsquointo the housersquo

L-CASES ADESSIVE vuoteella lsquoon the bedrsquo talolla lsquoat the housersquo

ABLATIVE vuoteelta lsquooff the bedrsquo talolta lsquofrom the housersquo

ALLATIVE vuoteelle lsquoonto the bedrsquo talolle lsquoto the housersquo

Table 3 The semantic opposition between s-cases and l-cases in Finnish

The six local cases are found in all Finnic languages except for most dialects of Livonian

where l-case endings are attested in non-productive relic forms only The extinct Salaca

dialect of Livonian had a set of productive l-cases which has sometimes been attributed to

2 Especially the traditional term lsquointernal local casesrsquo (Finnish sisaumlpaikallissijat) seems to be a misnomer as the

s-cases do not only signify a location lsquoinsidersquo or lsquoin the interior ofrsquo something Instead the s-cases in Finnic

languages can be seen as a semantically unmarked set of local cases as opposed to the l-cases signifying a

location in the exterior


63

Estonian influence (for different points of view on this see Sjoumlgren amp Wiedemann 1861 37ndash

38 72ndash74 Itkonen 1957a 310ndash311 Kettunen 1957 429ndash430 Itkonen 1957b 435ndash436)3

However no clear cognates to the Finnic l-cases are found in more distantly related Uralic

languages Mari and Permic languages also have cases built with a coaffix -l- but their

functions are possessive rather than local Even though the Finnic and Mari-Permic l-cases

have often been seen as historically related they have usually been considered the result of

convergent development hence no l-cases are normally reconstructed to the proto-language

common to Finnic Mari and Permic (ie Proto-Finno-Permic in the traditional taxonomical

scheme)4

In this study our aim is to examine the historical background of the Finnic l-cases

applying the received methods of comparative linguistics As will be shown below other

Uralic languages ndash especially Saami and Permic languages ndash yield decisive evidence of the

historical origins of these cases In addition we will also present some hypotheses of the

possible origins of the l-cases in Mari and Permic languages even though these are not the

main object of our study


Apparently the first scientific work in which Finnic l-cases have been compared to forms in

other Uralic languages is Rasmus Raskrsquos Saami grammar Raeligsonneret lappisk Sproglaeligre

efter den Sprogart som bruges af Fjaeligldlapperne i Porsangerfjorden i Finmarken (1832)

Rask equated the Finnic l-cases with the North Saami postpositions alde lsquoonrsquo and ala lsquoontorsquo

and also suggested that some North Saami adverbs built with the coaffix -l- (eg davil lsquofrom

northrsquo olggul lsquofrom outsidersquo) had developed from the same source

[Finsk]

Tilf[ormen] tograve l i l le panna laeliggge paring Stolen

Vedf[ormen] tograve l i l la istua sidde paring Stolen [ndash ndash]

Fraf[ormen] tograve l i lda ottaacute tage bort af Stolen

[ndash ndash]

3 In Karelian the allative (-lle) has rather recently coalesced with the adessive in -lla 4 The internal classification of Uralic languages is matter of ongoing dispute and currently there is no consensus

as to whether lsquoFinno-Permicrsquo forms a valid node within Uralic see eg Salminen (2002) for a critical view


64

[ndash ndash] Til Bevis at den ogsaring har vaeligret den oprindelige i Lappisk maring tjene [ndash ndash]

Tilf vare-ala op paring Bjaeligrget davvele mod Norden vaʒe olggole uwsa Garing udenfor Doumlren

Vedf vare-aldrsquo (aldn) paring Bjaeligrget davvelestrsquo nord paring olggolestrsquo uden for

Fraf vare-ald fra Bjaeligrget daveld norden fra olgold uden fra

(Rask 1832 35ndash36)

lsquo[Finnish]

to-f[orm] togravel i l le panna put on the chair

at-f[orm] togravel i l la istua sit on the chair [ndash ndash]

from-f[orm] togravel i lda ottaacute take off the chair

[ndash ndash]

[ndash ndash] For proof that this has also originally been the case in Saami serves [ndash ndash]

to-f vare-ala onto the mountain davvele northward vaʒe olggole uwsa go outside the door

at-f vare-aldrsquo (aldn) on the mountain davvelestrsquo in the north olggolestrsquo outside

from-f vare-ald from the mountain daveld from north olgold from outsidersquo

Rask based this comparison on his observations of the functional similarity between Finnic l-

cases and Saami al-postpositions He hypothetisized that Saami also had originally had a set

of l-cases but the case endings had split off the nouns and become independent words

retaining their original suffixal status in only certain adverbs

De naeligste tre Former have unaeliggtelig fundet Sted i Sproget som er indlysende af d a v v e l e

d a v v e l e s t rsquo d a v e l d men disse Endelser bruges nu som det synes kun i nogle gamle No der ere

ufuldstaeligndigen tilovers som blotte Forholdsord eller Biord f E b a ʒ j e ( p a ʒ j e ) som er oventil haves i

disse Former aldeles overensstemmende med det finske p ǽ ( p aelig aelig ) Hoved der ogsaring i de samme Former

bruges paring samme Maringde saringledes

Finsk Lappisk

Tilf pǽlle bagjele op over op paring

Vedf pǽllaelig bagjelestrsquo oven over oven paring

Fraf pǽldaelig bajeld ovenfra nedenfra

[ndash ndash] Men disse Endelser forekomme som sagt kun i nogle enkelte Ord i de fleste Tilfaeliglde ere de blevne

afrevne fra Ordene i en noget forskjellig Form og betragtede som saeligregne Forholdsord hos L[eem (1748)]

findes kun to saringdanne nl a l a hen paring til [ndash ndash] og a l d som L oversaeligtter paring [ndash ndash] (Rask 1832 37ndash38)


65

lsquoThe next three forms [= l-cases] have undeniably occurred in the [Saami] language which is obvious from

d a v v e l e d a v v e l e s t rsquo d a v e l d but these endings are now apparently only used in certain old

nouns that remain defective as bare adpositions or adverbs eg b a ʒ j e ( p a ʒ j e ) that which is above

occurs in these forms altogether analogous to Finnish p ǽ ( p aelig aelig ) head which is also used in the same

manner in the same forms thus

Finnish Saami

to-f pǽlle bagjele op over op paring

at-f pǽllaelig bagjelestrsquo oven over oven paring

from-f pǽldaelig bajeld from above down from

[ndash ndash] But as said these endings only occur in certain individual words in most cases they have split off from

words in a somewhat separate form and regarded as separate adpositions in L[eem (1748)] only two such

are found namely a l a onto [ndash ndash] and a l d which L[eem] translates as paring [lsquoonrsquo] [ndash ndash]rsquo

Raskrsquos explanation was apparently adhered to by M A Castreacuten in his doctoral dissertation

De affinitate declinationum in lingua Fennica Esthonica et Lapponica (1839) Castreacuten

accepted the equivalence of l-cases and Saami al-postpositions at least on a synchronic level

and seems to maintain that the morphemes are also etymologically cognate

Casus qui nominati sunt Allativus Adessivus Ablativus e lingua Lapponica omnino fere evanuerunt neque

occurrunt nisi in quibusdam adverbiis et praepositionibus ex gr bagje -le (Fenn paumlauml -lle Allat) baje -ld

(Fenn paumlauml -ltauml Ablat) siskele siskeld davvele davveld e s p Adessivus in illis quoque vocibus

compensatur Infinitivo Allativum nominum compensat postpositio ala Adessivum interdum aln (aldn

Rask) saepissime vero ald quae proprie post Ablativum ponitur[5] (Castreacuten 1839 59)

lsquoThe cases which were mentioned allative adessive ablative have altogether disappeared in the Saami

language and do not occur except in certain kinds of adverbs and prepositions [= postpositions] for

example bagje -le (Finn paumlauml -lle allat[ive]) baje -ld (Finn paumlauml-ltauml ablat[ive]) siskele siskeld davvele

davveld etc The adessive in those expressions is compensated for by the infinitive [= partitive] The allative

of nouns is compensated by the postposition ala the adessive sometimes with aln (aldn Rask) most often

however ald which is properly placed after the ablative[5]rsquo

Raskrsquos explanation was also supported by Stockfleth (1840 10) but after this the idea seems

to have sunk into oblivion In his later publications Castreacuten compared the Finnic l-cases to the

l-cases in Mari and Permic languages as well as to Khanty adverbs containing an element -l-

[5] E sect 28 apparet illas postpositiones primitus fuisse casuum terminationes [lsquoAccording to sect 28 it is clear that

these postpositions originally were case endingsrsquo such information cannot however be found in sect 28]


66

leaving Raskrsquos hypothesis entirely unmentioned (Castreacuten 1844 vi 17ndash22 1854 112ndash117

1858 [1849] 28) And already before this Loumlnnrot (1841 35ndash37) had proposed a different

explanation without making any reference to either Castreacuten (1839) or Rask he equated the

coaffix -l- with the Finnish word liki lsquonear almostrsquo and suggested that it had developed

through attrition from this lexical root the coaffix -s- in the endings of the s-cases he

explained on the basis of the root sisauml- lsquoinsidersquo

Other explanations based on relational nouns were also suggested in the latter half of

the 19th century Hunfalvy (1864 301) connected the coaffix -l- with the Finnic relational

noun luo- cf luona lsquoat (= in the vicinity of)rsquo luota lsquofrom (the vicinity of)rsquo luo ~ luokse lsquoto

(the vicinity of)rsquo On the other hand Ahlqvist (1863 26ndash27 1877 105ndash106) equated the -l-

with the Finnic root ala- lsquounder-rsquo Ahlqvistrsquos idea involved an interesting etymological

misunderstanding which brought it somehow close to Raskrsquos explanation he also maintained

that there is a relationship between Finnic l-cases and the Saami al-postpositions but he

mistakenly thought that the Saami postpositions were cognate with Finnish ala- lsquounderrsquo It is

true the regular vowel correspondences between Finnish and Saami were only later worked

out by Genetz (1896) but despite of this Sjoumlgren (1828 397) already had correctly analyzed

Saami al- as the cognate of Finnish ylauml- lsquoup above-rsquo instead But Ahlqvist thought the

Finnish forms talolla lsquoat the housersquo and talolta lsquofrom the housersquo were historically equivalent

to the North Saami expressions ldquodalo alardquo (= daacutelu ala) and ldquodalo aldrdquo (= daacutelu alde) in

reality though the latter two mean lsquoonto the housersquo and lsquoon the housersquo respectively Later

this mistaken equation of l-cases with Finnish ala- lsquounderrsquo was also supported by Blomstedt

(1869 44)

The early comparisons made by Loumlnnrot Hunfalvy and Ahlqvist have been recognized

by later research but on the other hand Donnerrsquos (1879 84ndash93) extensive discussion on the

relationships of l-cases and adverbs with an l-element in the Finno-Ugric languages seems to

have gone almost entirely unnoticed6 This is interesting as among the late 19th century

scholars Donner can be characterized as the only one who based their hypotheses concerning

the origin of the l-cases on a genuine comparative analysis Donnerrsquos treatment differs from

the earlier (and also most of the later) discussions on the l-cases in that he systematically tried

to show cognate forms between distantly related Finno-Ugric languages eg Finnish tuolta

6 As far as we are aware the only scholar who has referred to Donnerrsquos views on l-cases is Haumlkkinen (1984 7

9) who herself maintains that the system of l-cases would have developed in the Finnic-Saami proto-language

(lsquoEarly Proto-Finnicrsquo) already


67

lsquofrom therersquo ~ Khanty lttoltagt ( = Vakh Khanty taltǝɣ) id Finnish edellauml lsquoaheadrsquo ~ North

Saami ltauddalgt (= ovddal) lsquotowards (from the opposite direction)rsquo ~ Mari ltanzalnagt (=

West Mari anzǝlnǝ) lsquoaheadrsquo Finnish veneellauml lsquoat the boat by boat with a boatrsquo ~ Ter Saami

ltvanselgt lsquoby boat with a boatrsquo According to present knowledge most of these etymological

comparisons are erroneous though

Donner saw possible diachronic connections between the l-cases and several Finno-

Ugric word-roots such as the Finnic relational noun roots ala- lsquounderrsquo ete- lsquofrontrsquo ul-ko-

lsquooutsidersquo and uumlle- lsquoup aboversquo and their cognates In this connection Donner also mentions the

local derivational suffix -lA Donnerrsquos wordings are however rather cautious and in fact

difficult to interpret he does not take a clear stance as to which of these elements would lie

behind the formation of the l-cases but instead states that any one of them could account for

their origin

Wenn wir jetzt die frage uumlber den ursprung der l-kasus vom finnischen standpunkte zu beantworten suchen

so haben wir kein criterium um sie entweder mit ete vogul el magy el mit dem finnischen luo oder mit ala

in verbindung zu setzen Der bedeutung und der form nach koumlnnen sie aus allen hergeleitet werden [ndash ndash]

(Donner 1879 91)

lsquoIf we seek to answer the question of the origin of the l-cases from the point of view of Finnish then we do

not have a criterion for connecting them with either [Finnish] ete- Mansi el Hungarian el [lsquofront-rsquo] or

Finnish luo [lsquoat-rsquo] or ala [lsquounder-rsquo] According to form and meaning they can be derived from all of them [ndash

ndash]rsquo

Die bedeutung welche alle diese verschiedenen differenzirungen [= eg ala ete ul-ko uumlle] mit einander

verknuumlpft ist aussenseite flaumlche und haumlngt offenbar mit dem im finnischen und anderen sprachen

vorkommenden ableitungssuffix la zusammen welches lokalitaumlt wohnplatz aufenthaltsort bezeichnet und in

nahem zusammenhang mit luo naumlhe steht Aus dieser fruumlhen periode der sprachbildung leite ich daher den

ursprung der l-kasus durch ein suffix welches in naher beziehung zu allen den genannten steht (Donner

1879 92)

lsquoThe meaning that links all these varying differentiations [eg ala lsquounder-rsquo ete lsquofront-rsquo ul-ko lsquooutside-rsquo uumlle

lsquotop above-rsquo] with each other is lsquoexteriorrsquo lsquosurfacersquo and it is obviously connected with the derivational

suffix -lA occurring in Finnish and other languages which designates lsquolocalityrsquo lsquoplace of residencersquo

lsquowhereaboutsrsquo and which stands in a close connection with luo lsquonearrsquo From this early period of language

formation I derive the origin of the l-cases via a suffix which stands in a close relationship with all the

elements mentionedrsquo


68

After Donner a new phase began in the research history of the l-cases According to a new

view briefly presented by Budenz (1886 464) the l-cases would have their origin exactly in

the derivational suffix -lA that was already mentioned by Donner the suffix is attested in

such Finnish derivatives as eg appela lsquofather-in-lawrsquos housersquo (larr appi lsquofather-in-lawrsquo) and

pappila lsquoparsonagersquo (larr pappi lsquopastorrsquo) Budenz presented functional arguments for his

analysis in the form of two brief usage examples

Joacuteformaacuten egynek is vehető ezen -l keacutepzővel mellyel az emliacutetett casusok specialis tője alakuacutel a finn laquonomen

lociraquo-keacutepző -la -lauml pl appela domus soceri (appe) pappila domicilium sacerdotis (pappi) miehelauml dom

virorum mariti mert eredeti jelenteacutesűl raacuteillik a laquomelleacutek (mellette eacutes koumlruumlle valoacutesaacuteg)raquo-feacutele azt tekintve hogy

pl on miehelaumlssauml magyaruacutel iacutegy van laquofeacuterj n eacute l vanraquo meg mennauml miehelaumlhaumln (Kalev 23 496) laquofeacuterj h e z

menniraquo (Budenz 1886 464)

lsquoIn fact the derivational suffix -l that underlies the special stem of the cases mentioned [= l-cases] can be

equated with the Finnish laquonomen lociraquo derivative -la eg appela lsquofather-in-lawrsquos housersquo (appe [appi lsquofather-

in-lawrsquo]) pappila lsquoparsonagersquo (pappi [lsquopastorrsquo]) miehelauml lsquohusbandrsquos house marriagersquo [mies miehe- lsquoman

husbandrsquo] because something like ldquosupplementary lateralrdquo (ldquolocating or existing beside and aroundrdquo) suits

as the original meaning considering eg on miehelaumlssauml lsquofeacuterjneacutel van [manADE be3SG]rsquo [lsquois marriedrsquo]

mennauml miehelaumlhaumln (Kalevala 23 496) lsquofeacuterjhez menni [manALL goINF]rsquo [lsquoto get marriedrsquo]rsquo

Later Setaumllauml (1890) commented on the origin of the l-cases which he considered originating

from the derivative paumlaumlle- with a suffix -l(e)- (cf Finnish paumlaumll(l)ys lsquocover(ing)rsquo paumlaumll(l)ikkouml

lsquochief headrsquo paumlaumll(l)inen lsquocover upperrsquo) a similar idea had already been presented by

Loumlnnrot (1841) who maintained that the element -l(e)- was eventually a truncation of the

word liki lsquonear almostrsquo Setaumllauml refers to Budenzrsquos explanation and considers it possible that

the coaffix -l- is originally connected with the derivational suffix -lA he rejects explanations

based on postpositional stems Even so Setaumllaumlrsquos attitude is rather cautious

Suomalaiset muodot paumlaumlllauml lt paumlaumll-nauml paumlaumll-tauml ovat siis katsottavat vain l(e)-johtoisen sanan sijamuodoiksi

ja koko ulkoinen paikallisryhmauml on pidettaumlvauml taumlmmoumlisistauml johdannaisista alkunsa saaneena niin kuin

LOumlNNROT (Suomi 1841 5 v s 36) ja BUDENZ (aumlsken main p) ovat olettaneet Liian kauvas on menty kun

tahdotaan taumltauml laumlauml panna liki sanan yhteyteen (LOumlNNROT Suomi 1841 5 v s 37) tai johtaa sitauml ala sanasta

(AHLQVIST Suomi II 1 s 27 BLOMSTEDT Halotti Beszeacuted s 44) sitauml vastoin voisi sillauml ajatella olevan

yhteyttauml paikallisen la paumlaumltteen kanssa (vrt QVIGSTAD [1881] Beitr s [1]36 BUDENZ main p) (Setaumllauml

1890 409 emphasis added)


69

lsquoThe Finnish forms paumlaumlllauml lt paumlaumll-nauml paumlaumll-tauml must thus be seen merely as case forms of a word with a

derivational suffix l(e) and the whole group of external local cases considered originating from such

derivatives as Loumlnnrot (Suomi 1841 5 p 36) and Budenz (op cit) have assumed One has gone too far

when one has wanted to connect this l with the word liki [lsquonear almostrsquo] (LOumlNNROT Suomi 1841 5 p 37) or

to derive it from the word ala [lsquounder-rsquo] (AHLQVIST Suomi II 1 p 27 BLOMSTEDT Halotti Beszeacuted p 44)

instead one could think that it has some connection to the local suffix la (cf QVIGSTAD [1881] Beitr p

[1]36 BUDENZ op cit)rsquo

Doubts apparently vanished soon however Szinnyei (1910 73ndash75) presents the equation

with the suffix -lA laconically as if it were unanimously accepted

Im Ostseefinnischen im Tscheremissischen und in den permischen Sprachen gibt es eine Suffixgruppe

deren gemeinsames Element ein -l ist Dieses -l war urspruumlnglich ein Bildungssuffix und hat sich als solches

im Finnischen (-la -lauml) und in den permischen Sprachen (-la) bis jetzt erhalten z B finn pappila Pfarrhof

Pfarrhauslsquo (pappi Priesterlsquo) appela Haus des Schwiegervaterslsquo (appe-) anoppila Haus der

Schwiegermutterlsquo (anoppi) miehelauml- Haus des Manneslsquo (miehe-) [ndash ndash] (Szinnyei 1910 73ndash74)

lsquoIn the Finnic Mari and Permic languages there is a group of suffixes whose common element is -l This -l

was originally a derivational suffix and has been preserved as such in Finnish (-la -lauml) and in the Permic

languages (-la) eg Finnish pappila lsquoparsonagersquo (pappi lsquopastorrsquo) appela lsquofather-in-lawrsquos housersquo (appe-

[lsquofather-in-lawrsquo]) anoppila lsquomother-in-lawrsquos housersquo (anoppi [lsquomother-in-lawrsquo]) miehelauml- lsquohusbandrsquos housersquo

(miehe- [lsquoman husbandrsquo]) [ndash ndash]rsquo

Budenzrsquos explanation which we will henceforth call the lsquolA-theoryrsquo seems to have become

the commonly accepted view on the origin of the l-cases since then Wichmann (1913ndash1918

13ndash15) added another Finnic derivational type to the explanation namely cases where the

suffix -lA is attached to a relational noun root eg Finnish etelauml lsquosouthrsquo larr ete- lsquofrontrsquo (the

original meaning of etelauml was probably lsquoarea in front of the housersquo or the like as the front

sides of houses used to face south SSA sv etelauml) In such formations the derivational suffix

would supposedly have become reanalyzed as a part of a case ending because nouns referring

to a locality most often occur in local case forms Hakulinen (1941 90ndash91) mentions three

types of Finnic derivatives in connection with the lA-theory 1) derivatives based on relational

noun roots eg etelauml lsquosouthrsquo 2) oikonym derivatives eg appela lsquofather-in-lawrsquos housersquo

pappila lsquoparsonagersquo 3) derivatives based on pronoun roots and the suffix combination -kA-

lA- eg tauml- lsquothisrsquo rarr taumlkaumllauml rarr taumlkaumllaumlinen lsquolocal to this place inhabitant of this arearsquo taumlaumlllauml

lsquoherersquo (lt taumlkaumll-nauml) taumlaumlltauml lsquofrom herersquo (lt taumlkaumll-tauml)


70

Since the publication of Hakulinen (1941) these three types of derivatives have been

routinely mentioned in connection with the lA-theory and the explanation has become a piece

of textbook knowledge that is constantly referred to but practically never subjected to critical

discussion The theory has been described as lsquothe old and certainly correct viewrsquo (ldquovanha ja

varmasti oikea kaumlsitysrdquo Uotila 1945 334) ldquothe traditional viewrdquo (Tauli 1956 214) lsquothe

widespread commonly accepted hypothesisrsquo (ldquoраспространенная общепринятая

гипотезаrdquo Serebrennikov 1962 12 1963 47) and ldquothe accepted opinionrdquo (Anttila amp Uotila

1984 125) and since the 1930s it seems to have been accepted in nearly every publication in

which the origin of the Finnic l-cases has been commented upon7 But despite recurrent

expressions of support extremely little new evidence for the lA-theory has been presented

after Hakulinen The limited discussion on the issue has tended to concentrate on the

interrelations of the Finnic l-cases (with primarily local functions) and the Mari and Permic l-

cases (with primarily possessive functions) and they have usually been seen as results of

convergent development (eg Ravila 1958 13 Itkonen 1966 265ndash266 Reacutedei 1996 259ndash

260)8

In spite of its almost universal acceptance the lA-theory did not remain completely

without criticism The studies by Serebrennikov (1962 13 1963 47) are a notable exception

to the communis opinio He has paid attention to the fact that the semantics of the derivatives

in -lA or -l(V) are difficult to equate wit the functions of l-cases

Отсюда может быть сделан только один вывод элемент -l мог послужить показателем

внешнеместных падежей только в том случае если он сам обладал какой-то сходной семантикой

Насколько известно словообразовательный суффикс в таких образованиях как финск setaumllauml lsquoдом

дядиrsquo или коми-зыр бӧрла (дор) lsquoзадняя частьrsquo водзла (дор) lsquoпередняя частьrsquo такой семантикой не

обладает Поэтому если рассуждать чисто логически становится совершенно непонятно каким

образом этот элемент мог стать показателем внешнеместных падежей (Serebrennikov 1962 13

emphasis added)

7 In addition to the scholars already mentioned proponents of the lA-theory include at least Ravila (1935 43ndash45

1958 13) Lehtisalo (1936 148ndash150) Collinder (1952 11 1960 291) Tauli (1952 32ndash35) Pajusalu (1957a

159ndash160) Oinas (1961 8) Itkonen (1966 265ndash266) Raumltsep (1979 51ndash53) Korhonen (1979 9ndash10 1981 210ndash

211 231ndash232 1991 10) Laanest (1982 165ndash167) Haumlkkinen (1983 75ndash76 1985 86ndash87 2002 82ndash83) Baker

(1985 144) Leino (1990 126) Tikka (1992 40) Kulonen (1993 18ndash19 80ndash81) Huumo (1995 64ndash65) Reacutedei

(1996 259ndash260) Alhoniemi (2001) Kracht (2005) and Huumo amp Ojutkangas (2006 17) 8 Bartens (2000 82ndash83) however does not share this opinion her view will be discussed in more detail in

Section 42 below


71

lsquoHence only one conclusion can be drawn the element -l could serve as a marker of the external local cases

only in the case that the element itself possesses somehow similar semantics As far as is known the

derivational suffix in such formations as Finnish setaumllauml lsquounclersquos housersquo or Komi-Zyryan бӧрла (дор) lsquorear

sidersquo водзла (дор) lsquofront sidersquo does not possess such semantics Therefore if we think purely logically it

becomes entirely incomprehensible how this element could become a marker of the external local

casesrsquo

On the other hand Serebrennikov speculates that a Finno-Permic ldquosuperessiverdquo -l might lie

behind the l-cases but fails to present clear evidence for this hypothesis In spite of this

though his criticism quite clearly demonstrates the basic weakness of the lA-theory it simply

remains unexplained how the core functions of the Finnic (or the Mari-Permic) l-cases could

be connected with the semantics of the derivational suffix -lA The comparison seems to be

primarily based on mere similarity of form and the semantic relationship remains vague the

l-cases and the derivational suffix -lA show hardly any similarities of meaning beyond a

loosely defined ldquolocalrdquo function Even so Serebrennikovrsquos arguments have gained little

attention Apparently only Baker (1985) has tried to counter this criticism

Attempts to refute this theory [ndash ndash] by claiming an inflectional or postpositional source for the l morpheme

have foundered upon the formidable weight of derivational collateral provided by the contemporary

languages and the absence of independent comparative evidence to support the existence of an original

desinence or adposition featuring the l element which could reasonably have provided the base for some or

all of the cases (Baker 1985 144)

One must note that it remains quite unclear what the ldquoformidable weight of derivational

collateralrdquo mentioned by Baker is supposed to be (cf Baker 1985 144ndash153) It is true of

course that the Uralic languages possess a variety of words formed with some kind of ldquolocalrdquo

suffix or suffixes of the shape -l(V)- (and this is probably the case with many other language

families too) However as pointed out by Serebrennikov already the semantics of such

formations do not show any clear correspondence to the functions of the l-cases ndash and it

seems none of the supporters of the lA-theory have attempted to present a plausible account

of how this functional gap between the forms could be bridged Moreover to Serebrennikovrsquos

criticism one can add that the likelihood of chance resemblance is significantly increased by

the fact that the compared element consists of a single phoneme (l) which is moreover a

typologically common and unmarked sound and in such a case particularly strong arguments

are required for an etymological equation to be established


72

Baker is quite right though in noting that so far there have not been any plausible

attempts to equate the Finnic or the Mari-Permic l-cases with postpositions or other

grammatical elements In the next section we will show however that strong evidence for the

postpositional origin of the l-cases can be found


The analysis of previous research has revealed that the lA-theory despite of being generally

accepted has in fact never been substantiated with convincing semantic and functional

arguments This provides us a motive to approach the problem of the origin of the Finnic l-

cases from a quite different perspective In this study the received methods of comparative

linguistics form our methodological framework and particular attention will be paid to the

functions of cases We will seek to first identify the historically primary semantic function of

the Finnic l-cases and then to compare the cases to those structures in other Uralic languages

that exhibit the same semantic function The benefit of such an approach lies in its potential to

provide an answer to two distinct questions it may both reveal potential but so far undetected

cognate morphemes for the Finnic l-case suffixes and yield more information on the

grammatical expressions of external locality prior to the development of the l-cases in Pre-

Proto-Finnic and even in Proto-Uralic Indeed it can be said that the weakness of the

prevailing view ultimately stems from the lack of such a comparative approach the essence

of the lA-theory is formed by arguments supporting the equation of the coaffix -l- with the

derivational suffix -lA and it has never been expanded to include a detailed model of the

development of expressions of external locality from Proto-Uralic to Proto-Finnic

We will argue below that Rask (1832 37ndash38) is the only scholar who has come close to

the right solution of the problem Of course Raskrsquos idea of original l-case endings developing

into independent al-postpositions in Saami is erroneous in light of current knowledge of the

etymology of these postpositions Nevertheless the basic assumption of a diachronic

connection between Finnic l-cases and Saami al-postpositions is well motivated as the two

elements are not only similar in form but also show obvious functional affinities Hence we

can modify Raskrsquos explanation and postulate the hypothesis that l-case suffixes were

grammaticalized from earlier postpositions that were retained in Saami In what follows we

seek to verify this hypothesis through a detailed comparative analysis As the first step the

primary functions of the l-cases will be examined in more detail


73


The functions of l-cases in Finnish have been explained in detail by eg Alhoniemi (1979)

Leino (1989 1990) and Huumo (1995) and Estonian l-cases have been treated by Vainik

(1995) for discussion on the functions of l-cases in Finnic languages in general see Pajusalu

(1957b 1958a 1958b 1960) The core function of Finnic l-cases is to express location in the

proximity the vicinity and especially on the upper surface of something This has been aptly

put by Alhoniemi

l-sijaa kaumlytettaumlessauml puhutaan paikallissijaisten sanojen tarkoitteista tai tarkoitteiden pinnoista ikaumlaumln kuin

kaksi ulottuvuutta omaavina lokaliteetteina kun taas naumliden sanojen vastaavia sisaumlpaikallissijoja

kaumlytettaumlessauml tarkoitteet naumlhdaumlaumln kolmiulotteisina olioina joille on ominaista mm tilavuus ja materia Naumlin

siis Varissuolla ilmaisee kaksiulotteisen lokaliteetin kun taas Varissuossa-ilmausta kaumlytettaumlessauml suo naumlhdaumlaumln

myoumls syvyyttauml omaavana elementtinauml Vastaavalla tavalla vuoteelle matolle kadulle poumlydaumllle ilmoittavat

subjektin tai objektin tarkoitteen olinpaikaksi pelkaumln pinnan kun taas vastaavat sisaumliset paikallissijat

suhteuttavat tarkoitteet kolmiulotteiseen maailmaan (Alhoniemi 1979 94)

lsquoWhen using an l-case one speaks of the referents and their surfaces as if they were localities involving two

dimensions whereas when using a corresponding internal local case [s-case] the referents are seen as three-

dimensional objects which are characterized by eg volume and material Thus Varissuolla [crow-bog-

ADE] expresses a two-dimensional locality [lsquoon Crowbogrsquo] whereas when the expression Varissuossa [crow-

bog-INE] is used the bog is perceived as an element that also possesses depth [lsquoin Crowbogrsquo] In the same

way vuoteelle [bed-ALL] matolle [carpet-ALL] kadulle [street-ALL] and poumlydaumllle [table-ALL] express the

location of the subjectrsquos or objectrsquos referent merely in terms of a surface whereas the corresponding

internal local cases proportion the referents to the three-dimensional worldrsquo

Such usage can be shown as primary on levels of both synchronic description and diachronic

reconstruction In addition to these strictly local functions the main functions of l-cases

include possessive use as well as instrumental use of the adessive case However only the

local functions are fully shared by the Finnic languages Possessive use is missing in Livonian

(except for the Salaca dialect whose l-cases may result from Estonian influence) and even

across other Finnic languages possessive use is somewhat heterogeneous (Inaba 2001) which

suggests its secondary origin (see 42 for further discussion) The instrumental use of the

adessive in turn is characteristic of northern Finnic languages only Laaksonen (2000) has

compared the use of the adessive case in Finnish and Estonian and found the correspondence

to be highest in local functions especially in the so-called ON-function (ie lsquolocation on the


74

upper surfacersquo) Hence it is not surprising that this function has also been considered

diachronically primary as summed up by Vainik for instance

l-kaumlaumlnded kui ajalooliselt hilisemad peavad ju olema kasutusele votildeetud mingi markeeritud situatsiooni tarvis

Eeldades et areng toimub ikka konkreetselt abstraktsele tuleb arvata et l-kaumlaumlnete kasutuselevotildetu ajal oli

selleks markeeritud olukorraks totildeenaumloliselt pidepunkti 2-motildeotildetmelisus ja aluse funktsioon kui kotildeige

konkreetsem ja saumltestatum VK-dega taumlhistatav suhe (Vainik 1995 146 emphasis in the original)

lsquoThe diachronically more recent l-cases must have been taken into use for the needs of some kind of marked

situation Presupposing that development always takes place from the concrete to the abstract one can

believe that at the time when l-cases were taken into use that marked situation was probably the two-

dimensional nature of the point of reference and the function of underlying surface as it is the most

concrete and established relationship signified by the external local casesrsquo

Considering these findings the study of the origin of the l-cases naturally must begin by

examining what structures other Uralic languages use to express the same semantic function

ie lsquolocation on the upper surfacersquo In fact we find it quite odd that this crucial question has

almost never been addressed in previous studies on the subject

The comparative method shows quite unambiguously that Proto-Uralic had a series of

local postpositions formed from the relational noun root uumll(i)- lsquoplace up or aboversquo uumll-nauml

lsquoon-LOCrsquo uumll-tauml lsquoon-ABLrsquo and uumlli-ŋ lsquoon-LATrsquo These postpositions have retained their

primary functions in Saami Permic and Samoyed languages and they are also reflected in a

semantically slightly different Finnish series of adverbs and postpositions yllauml lsquoaboversquo yltauml

lsquofrom aboversquo and ylle lsquoto aboversquo As the reconstructed Proto-Uralic uumll-postpositions bear a

close resemblance to the endings of the l-cases and the two share the same semantic function

(see Table 4) the hypothesis that l-cases have emerged through agglutination of these

postpositions appears very promising indeed


75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish

uumll-nauml ńińa vilin vilin alne (alde) nanna -llA

uumll-tauml ńid˚ (viliś) (viliś) (alne) alde nalta -ltA

uumlli-ŋ ńih vile vile oolacirc ala nali -lle

Table 4 The reflexes of Proto-Uralic uumll-postpositions in some Uralic languages The forms put in parentheses

are functionally equivalent but not morphologically cognate with the other items in the row

Prototypical examples of the core local functions inherited from Proto-Uralic can be seen in

the following Komi (1andash5a) and Tundra Nenets (6andash8a)10 sentences the examples derive

from Reacutedeirsquos (1962) and Mikolarsquos (1975) studies on postpositions in Komi and Nenets

respectively As our translations of these sentences into North Saami (1bndash8b) and Finnish

(1cndash8c) reveal there is quite a clear correspondence between Finnish l-cases and the reflexes

of the Uralic uumll-postpositions in the core local functions

9 The Tundra Nenets ńi- postpositions are indeed cognate with Saami (n)al- and Permic vil- postpositions

despite the phonological dissimilarity The Proto-Samoyed forms of the postpositions are reconstructed as i-nauml

lsquoon-LOCrsquo i-tə lsquoon-ABLrsquo and i-ŋ lsquoon-LATrsquo The root i- has developed from earlier ij- lt uumlj- lt uumlľ- (lt Proto-

Uralic uumlli-) The nasal prothesis in Nenets (i- gt ŋi- gt ńi-) is a regular sound change The etymology and

phonological development of the Samoyed root i- is discussed by Janhunen (1981 256) To his discussion we

can add a possible explanation of the unexpected lack of the reflex of the lateral l In Uralic i-stems the lateral

was regularly palatalized and developed into the semivowel j in Samoyed and consequently the expected

reflex of the Proto-Uralic root uumlli- is Proto-Samoyed ij- Janhunen suggests that the irregular reduction of ij-

to i- could have been caused by lack of stress due to frequent use of the root in postpositions This is

conceivable but another explanation can also be proposed The locative and ablative forms which also

functioned as postpositions appear to have been formed from consonant stems (Proto-Uralic uumll-nauml and uumll-tauml

respectively) and it may well be that the loss of the lateral l is regular before the apical consonants n and t

There seems to be at least one parallel example of the development lt gt t in Samoyed namely Proto-Samoyed

karingtaring- lsquokillrsquo lt Proto-Uralic kal-ta- (a causative derived from the consonant stem of the verb kali- lsquodiersquo) so the

development of Proto-Uralic uumll-tauml to Proto-Samoyed i-tə can be interpreted as regular No other examples of

the Proto-Uralic cluster ln are known but since such a cluster does not seem to occur in Samoyed the

regularity of the development uumll-nauml gt i-nauml seems at least a valid possibility Thus the anomalous root form i-

(instead of ij-) could have been analogically generalized from the two forms reflecting Proto-Uralic consonant

stem formations 10 We are obliged to Tapani Salminen for converting the Tundra Nenets examples into phonological

transcription


76

(1) a akań kujle Nasťa ki vilin

doll lie3SG N hand vilin

b lsquodohkkaacute lea Nastja gieđa aldersquo

doll be3SG NGA handGA alde

c lsquonukke on Nastjan kaumldellaumlrsquo

doll be3SG NGEN handADE

lsquoThe doll is lying on Nastjarsquos handrsquo (Reacutedei 1962 15)

(2) a me tajes viľpev leććeda ju vile

1SG thisACC again take1SG river vile

b lsquodoalvvun daacuten ođđasit joga alarsquo

take1SG thisGA again riverGA ala

c lsquovien taumlmaumln uudestaan joellersquo

take1SG thisGEN again riverALL

lsquoI will take this on the river againrsquo (Reacutedei 1962 18)

(3) a bi vilin pert ešale

fire vilin cauldron hang3SG

b lsquodola alde heaŋgaacute ruitursquo

fireGA alde hang3SG cauldron

c lsquotulella riippuu patarsquo

fireADE hang3SG cauldron

lsquoThere is a cauldron hanging over the firersquo (Reacutedei 1962 14)

(4) a tuj vilin caŕ medis mužikliś juaśni

road vilin tsar beginPST3SG manABL askINF

b lsquogeainnu alde caacutera aacutelggii jearahallat aacutedjaacutesrsquo

roadGA alde tsar beginPST3SG askINF manLOC

c lsquotiellauml tsaari kaumlvi tiedustelemaan ukoltarsquo

roadADE tsar beginPST3SG askINF manABL

lsquoOn the road the tsar began to ask the old manrsquo (Reacutedei 1962 16)

(5) a a pizan vilin ńi-nem abu

but table vilin no-one NEGEX


77

b lsquomuhto beavddi alde ii leat mihkkegersquo

but tableGA alde NEG3SG beCNG nothing

c lsquomutta poumlydaumlllauml ei ole mitaumlaumlnrsquo

but tableADE NEG3SG beCNG nothingPTV

lsquoBut on the table there is nothingrsquo (Reacutedei 1962 15)

(6) a num ńīśawdeg solotejdeg toldeg‿‿‿‿ńińa ŋamťowideg

n father1SG golden throne(GEN)‿‿‿‿ńińa sit-INFR

b lsquoaacutehččaacuten Num čohkkaacutei golletruvnnu aldersquo

father1SG N sitPST3SG goldenthroneGA alde

c lsquoisaumlni Num istui kultaisella valtaistuimellarsquo

father1SG N sitPST3SG goldenADE throneADE

lsquoMy father Num sat on a golden thronersquo11 (Mikola 1975 48)

(7) a ŋardegwendeg ńińa məńdeg jaddegəmdegh

toeGEN1SG ńińa 1SG walk1SG

b lsquovaacuteccaacuten juolgesuorpmaid aldersquo

walk1SG toePLGA alde

c lsquokaumlvelen varpaillanirsquo

walk1SG toePLADE1SG

lsquoI walk on my toesrsquo (Mikola 1975 48)

(8) a serdeg‿‿‿‿ńin‿‿‿‿təneydegńih

iceGEN‿ńih‿step1DU

b lsquomoai laacutevkiime jieŋa alarsquo

1DU stepPST1DU iceGA ala

c lsquoastuimme jaumlaumlllersquo

stepPST1PL iceALL

lsquoWe stepped on the icersquo (Mikola 1975 46)

As the Finnic l-cases and Uralic uumll-postpositions show both functionally and phonologically

such a transparent correspondence it is quite surprising that very little attention has been paid 11 Num is the name of the Nenets supreme deity


78

to this since Rask (1832) The fact has not gone completely unnoticed during the period of the

lA-theory however Leino (1990) and Tikka (1992) suggest that the Finnish postpositional

series yllauml yltauml ylle lsquoaboversquo ndash and peculiarly also alla alta alle lsquobelowrsquo ndash could have had a

semantic influence on the development of the l-cases Even so they consider derivatives

based on the suffix -lA as the primary material source of the case forms

Here internal reconstruction leads to the conclusion that precisely those C-predicates that profile vertical

spatial relations ie the alla and yllauml sets of p-positions may have strongly influenced the development of

the l-cases [ndash ndash] The alla and yllauml sets seem to have offered a motivation for the fact that the l-cases acquired

the meaning lsquotop surface contactrsquo and thus are closely associated with the vertical dimension (Leino 1990

138ndash139 Footnote 12)

Koska naumlmauml ndash kuten edellauml on tullut esille ndash kuuluvat laumlhitienoita luotaaviin postpositioihin ei ole lainkaan

mahdotonta ettauml ne ovat olleet edesauttamassa ulkoisten paikallissijojen synnyssauml (Tikka 1992 40)

lsquoBecause these [ie the alla and yllauml sets] ndash as was noted earlier ndash belong to postpositions charting the

immediate vicinity it is not at all impossible that they have contributed to the birth of the external local

casesrsquo

One should note that Leino and Tikka are in fact the only scholars subscribing to the lA-

theory who have ever even tried to explain how the l-cases acquired the function of lsquolocation

on the upper surfacersquo However their explanation can be significantly simplified by assuming

that the yllauml set of postpositions is the concrete source of the l-case forms and not a mere

semantic catalyst in their development


Even though the similarity between Finnic l-cases and Uralic uumll-postpositions is striking on a

superficial examination more detailed proof is naturally needed in order to establish their

historical connection In this subsection an empirical test is performed we will examine how

and to what extent the use of the North Saami reflexes of Uralic uumll-postpositions

corresponds to the use of Finnic l-cases As seen in Table 4 (see Section 31 above) North

Saami has two postpositions inherited from the Uralic uumll-set namely a directional

postposition ala lsquoontorsquo and the postposition alde which has both a locative function (lsquoonrsquo) and


79

a separative function (lsquooff fromrsquo)12 Hence our hypothesis predicts that the functions of

North Saami ala will show a systematic resemblance to those of the Finnish allative case and

the functions of alde to those of the adessive and ablative cases

In order to test the hypothesis we have made use of a North Saami text corpus

consisting of 12 works of fiction four non-fiction titles the translation of The New Testament

published in 1998 and approximately 150 issues of the newspaper Min Aacuteigi from the years

1995 and 1997 (see the references for more details) The size of the corpus is over a million

words and it includes 2031 tokens of the words alde and ala ndash 1963 instances of

postpositions and 68 of adverbs

Even though we have chosen North Saami as the sole representative of the Saami

languages in our analysis this is highly unlikely to cause any serious bias in the results

becuase the usage of cognate postpositions in other Saami languages does not seem to differ

much from North Saami For example on the basis of our own acquired L2 intuition it is clear

that Inari Saami alne oolacirc and Skolt Saami acircacutelnn ool are used in a manner highly similar to

North Saami alde and ala and this intuitive judgment is confirmed by an examination of texts

in these languages (eg IK Sammallahti 2004 2012) A more detailed study might of course

still reveal some minor statistical differences


Before a more detailed quantitative investigation it is worth while to take a brief qualitative

look at the material through a few selected examples In the examples below we have

provided the Saami sentences with both Finnish and English translations in order to illustrate

the functional correspondences between the Saami al-postpositions and various Finnish

constructions Unless otherwise mentioned all translations are our own A part of the

observations on the use of North Saami al-postpositions have already been published Ylikoski

(2006)

12 The grammatical distinction between locative and separative forms has been lost in North Saami as well as in

all Eastern Saami languages Originally the distinction was lost in the sigular forms of local cases due to sound

change the Proto-Saami inessive singular ending (-snē) and elative singular ending (-stē) merged into -s(t)

Subsequently the distinction was analogically obliterated also in the locative plural as well as in adverbs and

postpositions (Korhonen 1981 223ndash224 Sammallahti 1998 66ndash67) Etymologically North Saami alde lsquoon offrsquo

reflects the Uralic separative form uumll-tauml whereas in Eastern Saami the corresponding locative form was

generalized instead (cf Inari Saami alne Skolt Saami acircacutelnn lsquoon offrsquo lt Uralic uumll-nauml)


80

In the material the great majority of al-postpositions are used in a local function In a

typical case the postpositional phrase expresses location on the upper surface of a referent ndash

eg an artifact (9) (10) a natural place (11) (12) a natural object (13) (14) or a part of the

body (15) (16) In such cases the most natural Finnish translation for the phrase often

involves an l-case form as in the following quite prototypical examples13

(9) Nohkadeimmet filttiin ja raacutenuin duolji nalde

fallasleepPST3PL blanketPLLOC and quiltPLLOC hideGA alde

guolgabolsttar oaivve vuolde

furpillow headGA under

lsquoNukahdimme vilteissauml ja raanuissa taljalla

fallasleepPST3PL blanketPLINE and quiltPLINE hideADE

karvatyyny paumlaumln allarsquo

furpillow headGEN under

lsquoWe fell asleep [tucked] in blankets and quilts on a hide with a fur pillow under our

headsrsquo (Blind 1992 59)

(10) Na dan aacutehkus leai nieiddaš čohkkame aacuteiddi alde

well itGA oldwomanLOC bePST3SG girlDIM sitPROG fenceGA alde

lsquoNo sillauml eukolla oli pieni tyttouml istumassa aidallarsquo

well itADE oldwomanADE bePST3SG little girl sitPROG fenceADE

lsquoWell that old woman had a little girl sitting on the fencersquo (Turi 1982 91)

(11) Maacutenaacutet ieža goivo alcceseaset jieŋa ala skeittaacutensaji

childPL REFLPL digPST3PL REFLILL3PL iceGA ala skatingplaceGA

lsquoLapset itse kaivoivat itselleen jaumlaumllle luistelupaikanrsquo

childPL REFL digPST3PL REFLALL3PL iceADE skatingplaceGEN

lsquoThe children themselves dug a skating place for themselves on the icersquo (MAacute 1995)

13 The examples also illustrate the variant forms in which these postpositions appear in North Saami alde ~ al ~

nalde and ala ~ nala The form al is merely an irregularly eroded form that is very common in spoken language

but less frequent in literary use The origin of the secondary initial nasal in nalde and nala is explained in

Section 36 and phonological erosion of these postpositions is discussed in more detail in Sections 36 and 37


81

(12) [ndash ndash] su ovddal bođii baacutelga al okta boares aacutehkku

3SGGA against comePST3SG pathGA alde one oldATTR woman

lsquo[ndash ndash] haumlntauml vastaan tuli polulla yksi vanha eukkorsquo

3SGPTV against comePST3SG pathADE one old woman

lsquo[ndash ndash] an old woman came towards him on the pathrsquo (Aikio amp Aikio 1978a 100

1978b 100)

(13) [ndash ndash] oainnaacuten mieđabealde muhtin geađggi alde goaskima čohkohaddamin

see1SG underthewind some rockGA alde eagleGA sitPROG

lsquo[ndash ndash] naumlen kotkan istuksivan tuulen alla jollain kivellaumlrsquo

see1SG eagleGEN sitINF windGEN under someADE rockADE

lsquo[ndash ndash] I see an eagle sitting on some rock under the windrsquo (Sombi 1996 8)

(14) Aacutehčči lebbii gaacutertta saacuteddo ala [ndash ndash]

father spreadPST3SG mapGA sandGA ala

lsquoIsauml levitti kartan hiekalle [ndash ndash]rsquo

father spreadPST3SG mapGEN sandALL

lsquoFather spread out the map on the sand [ndash ndash]rsquo (Jansson 1990 24 1979 24)

(15) De bajidii son su gieđas mu oalggi ala

then raisePST3SG 3SG 3SGGA handGA3SG 1SGGA shoulderGA ala

lsquo[Sitten] haumln nosti kaumltensauml olkapaumlaumlllenirsquo

[then] 3SG raisePST3SG handGEN3SG shoulderALL1SG

lsquoThen he raised his hand on my shoulderrsquo (Haeligtta amp Baeligr 1982 113 1993 153)

(16) Geasset saacutehtii bidjat cuoppolastta haacutevi nala

insummer bepossiblePST3SG putINF pondweedGA woundGA ala

lsquoKesaumlllauml saattoi laittaa uistinvidan lehden haavallersquo

summerADE bepossiblePST3SG putINF pondweedGEN leafGEN woundALL

lsquoIn summer one could put a pondweed leaf on a wound (in order to disinfect it)rsquo (Blind

1992 83)

Occasionally the actual function of the al-phrase is not really local but involves a transparent

metaphor based on a local meaning as in the following cases


82

(17) [ndash ndash] de baacutehcaacute buot bargu daacuteppe Rainer haacuterduid ala

then remain3SG all work here RainerGA shoulderPLGA ala

lsquo[ndash ndash] sitten jaumlauml kaikki tyouml taumlaumlllauml Rainerin harteillersquo

then remain3SG all work here RainerGEN shoulderPLALL

lsquo[ndash ndash] then all work here is left as Rainerrsquos responsibility (ldquoon Rainerrsquos shouldersrdquo)rsquo

(MAacute 1995)

(18) Dan vuođu ala mii saacutehttit hukset boahtteaacuteiggi

itGA foundationGA ala 1PL can1PL buildINF futureGA

lsquoSille pohjalle voimme rakentaa tulevaisuuttarsquo

itALL foundationALL can1PL buildINF futurePTV

lsquoOn that foundation we can build the futurersquo (MAacute 1995)

Some cases involve a referent which lacks a concrete upper surface or top (19) In the case of

body parts the phrase most often expresses posture (20) similar use occasionally occurs with

inanimate objects as well (21) Even in such cases the phrase can often be translated with an l-

case form

(19) Na de olmmaacutei vaacutelddii ja suddadii laju dola nalde [ndash ndash]

well then man takePST3SG and meltPST3SG leadGA fireGA alde

lsquoNo sitten mies otti ja sulatti lyijyauml tulella [ndash ndash]rsquo

well then man takePST3SG and meltPST3SG leadPTV fireADE

lsquoThen the man took and melted lead on the fire [ndash ndash]rsquo (Blind 1992 120)

(20) Baacutehppa Stockfleth maidda čohkka muohttat alde čippiid alde [ndash ndash]

pastor Stockfleth also sit3SG snowGA alde kneePLGA alde

lsquoPappi Stockfleth myoumls istuu lumella polvillaan [ndash ndash]rsquo14

pastor Stockfleth also sit3SG snowADE kneePLADE3SG

lsquoPastor Stockfleth also sits on the snow on his knees [ndash ndash]rsquo (Haeligtta amp Baeligr 1982 53)

14 The passage in the published Finnish translation is not an exact equivalent Pappi Stockfleth polvisteli muiden

tavoin lumella [ndash ndash] lsquoFather Stockfleth sat like the others kneeling on the snowrsquo (Haeligtta amp Baeligr 1993 88ndash89)


83

(21) [ndash ndash] biila lea fierran moddii birra ovdal bisaacutenii fas

car be3SG rollPSTPTCP afewtimes around before stopPST3SG again

juvllaid nala

wheelPLGA ala

lsquo[ndash ndash] auto on pyoumlraumlhtaumlnyt muutaman kerran ympaumlri ennen kuin

car be3SG rollPSTPTCP afewGEN occasionGEN around before than

pysaumlhtyi taas renkailleenrsquo

stopPST3SG again wheelPLALL3SG

lsquo[ndash ndash] the car has rolled over a few times before stopping on its wheels againrsquo (MAacute

1995)

Another type of semantic extension is the occasional use of Saami al-postpositions to

designate a location not lsquoon (the upper surface)rsquo but merely next to or in the immediate

vicinity of the referent These kinds of examples resemble the use of Finnish l-cases in the AT-

function (eg Finnish talolla lsquoat the housersquo) Hence they are often naturally translated with l-

case forms as in the case of (22ndash24) below One can compare (22) and (23) against (4) and

(2) in Section 31 in which the phrases geainnu alde lsquoon the roadrsquo and joga ala lsquoonto the

riverrsquo appear in a more prototypical ON-function

(22) Bargostohpu leai min skuvlageainnu nalde

workshop bePST3SG 1PLGA schoolwayGA alde

lsquoTyoumlpaja oli meidaumln koulutiellaumlmmersquo

workshop bePST3SG 1PLGEN schoolwayADE1PL

lsquoThe workshop was along our way to schoolrsquo (Blind 1992 71)

(23) Dainna mielain son vulggii Giru gillaacutei Avviljoga ala [ndashndash]

itCOM mindCOM 3SG leavePST3SG GiruGA villageILL AvviljohkaGA ala

lsquoSillauml mielellauml haumln laumlhti Kyroumln kylaumlaumln Ivalojoelle [ndash ndash]rsquo

itADE mindADE 3SG leavePST3SG KyroumlGEN villageILL IvalojokiALL

lsquoIn that mood he left for the village of Giru along the river Avviljohka [ndash ndash]rsquo (Castreacuten

2005 27)


84

(24) Dan botta skihpaacuterat ledje joavdan unna

itGA whileGA companionPL bePST3PL arrivePSTPTCP smallATTR

aacutedjagačča ala [ndash ndash]

brookDIMGA ala

lsquoSillauml vaumllin kumppanit olivat saapuneet pienelle purolle [ndash ndash]rsquo

itADE while companionPL bePST3PL arrivePSTPTCPPL smallALL brookALL

lsquoMeanwhile the companions had come upon a small brook [ndash ndash]rsquo (Castreacuten 2005 22)

As (9)ndash(24) illustrate the correspondences between Saami al-postpositions and Finnish l-

cases are rather pervasive in local functions Even so there are of course also many instances

where the Saami postpositional phrases cannot despite of having a local function be

translated with a Finnish l-case form As pointed out by Lauranto (1994 49) Finnish l-cases

are usually used in local functions only if the referent of the noun has a prominent upper

surface either in terms of the referentrsquos form or its function The local semantics of Saami al-

postpositions are stronger and hence their use is not as strictly limited by the nature of the

referent of the complement of the postposition For instance the following examples involve

referents that either have an upper surface that is not central to the function of the referent (a

car [25]) or lack a coherent upper surface altogether (gas bottles [26]) In such cases it is more

natural or even necessary to translate the Saami al-postposition with a Finnish postposition of

the series paumlauml-llauml lsquoon-ADErsquo paumlauml-ltauml lsquoon-ABLrsquo paumlauml-lle lsquoon-ALLrsquo15

(25) Faacutehkka almmaacutei njuikii eret biilla alde [ndash ndash]

suddenly man jumpPST3SG away carGA alde

lsquoYhtaumlkkiauml mies hyppaumlsi auton paumlaumlltauml [ndash ndash]rsquo

suddenly man jumpPST3SG carGEN paumlaumlltauml

lsquoSuddenly the man jumped off the car [ndash ndash]rsquo (Marastat 1990 28)

(26) Son bajidii guoros gaacutessa mii lei gaacutessaboahtaliid alde

3SG raisePST3SG emptyATTR boxGA which bePST3SG gasbottlePLGA alde

lsquoHaumln nosti tyhjaumlauml laatikkoa joka oli kaasupullojen

3SG raisePST3SG emptyPTV boxPTV which bePST3SG gasbottlePLGEN

paumlaumlllauml [ndash ndash]rsquo

15 This historical background of this postpositional series is discussed in more detail in Section 5


85

paumlaumlllauml

lsquoHe raised an empty box that was [placed] on the gas bottles [ndash ndash]rsquo (Jansson 1990 195

1979 175)

In the material one can also observe other types of limitations to the use of l-cases in a local

function In al-phrases that involve animate referents the translation to a Finnish l-case form

is impossible not only because of the lack of a coherent lsquoupper surfacersquo but also due to the

fact that the l-cases have possessive and dative functions associated with animate referents

Consider the following example

(27) Nisu gii gohčoduvvo bolesiin boares oahpisin njoarai

woman who callPASS3SG policePLLOC oldATTR acquaintanceESS pourPST3SG

godena olbmaacute nala ja cahkkehii su

moonshineGA manGA ala and ignitePST3SG 3SGGA

lsquoNainen jota poliisit kutsuvat vanhaksi tutuksi

woman whichPTV policePL call3PL oldTRANSL acquaintanceTRANSL

kaatoi pontikkaa miehen paumlaumllle ja sytytti haumlnetrsquo

pourPST3SG moonshinePTV manGEN paumlaumllle and ignitePST3SG 3SGACC

lsquoA woman who is called an old acquaintance by the police poured moonshine on the

man and set him on firersquo (MAacute 1995)

In (27) there is simply no possibility of translating Saami ala with the Finnish allative case

as in connection with a human referent the case has a dative function it is naturally

something altogether different to lsquopour the man some moonshinersquo (kaataa pontikkaa

miehelle) than to lsquopour moonshine on the manrsquo (kaataa pontikkaa miehen paumlaumllle) But it

should be noted that such restrictions to the local use of l-cases have become necessary only

when the possessive functions of these cases have first started to develop in Finnic As

already mentioned in 31 the possessive use must be interpreted as secondary its

development will be discussed in more detail in Section 42 below

In addition to the core local functions al-postpositions are also encountered in various

kinds of other uses In particular alde and ala often occur as postpositions governed by a

variety of verbs These kinds of cases make up over one tenth of the sentences in the research

material and their Finnish equivalents are quite heterogeneous In the following examples

the verbs doarrut lsquoto fightrsquo (28) suhttat lsquoto get madrsquo (29) and jurddahit lsquoto thinkrsquo (30)


86

govern either an alde phrase or an ala phrase such usage is analyzed in more detail by

Ylikoski (2006)

(28) Ahte vel moaluid alde nai daacuterbbašit doarrut daacutekkaacuter rikkis

that still crumbPLGA alde also need3PL fightINF thiskindof rich

stuorra gaacutevpogis

largeATTR cityLOC

lsquoEttauml heidaumln vielauml muruista=kin tarvitsee tapella

COMP 3PLGEN still crumbPLELA=also need3SG fightINF

taumlllaisessa rikkaassa suuressa kaupungissarsquo

thiskindofINE richINE bigINE cityINE

lsquo[Imagine] that they must fight even for crumbs in such a rich and large cityrsquo (Vars

1990 46)

(29) [ndash ndash] Ovllaacute-viellja meinnii duođas suhttat mu ala

Ovllaacutebrother beabouttoPST3SG seriously getmadINF 1SGGA ala

vaikko mun in dadjan maidige (Vest 1988 28)

eventhough 1SG NEG1SG sayCNGPST nothingACC

lsquo[ndash ndash] Oula-veli meinasi tosissaan suuttua minulle

Oulabrother beabouttoPST3SG seriously getmadINF 1SGALL

vaikka minauml en sanonut mitaumlaumlnrsquo

eventhough 1SG NEG1SG sayCNGPST nothingPTV

lsquoBrother Ovllaacute almost got mad at me for real even though I didnrsquot say anythingrsquo (Vest

1990 28)

(30) In mon gal jurddahan ruđa nala mon ledjen dalle ain nu

NEG1SG 1SG really thinkCNGPST moneyGA ala 1SG bePST1SG then still so

maacutenas

childish

lsquoEn minauml kyllauml ajatellut rahaa olin silloin vielauml niin

NEG1SG 1SG really thinkCNGPST moneyPTV bePST1SG then still so

lapsellinenrsquo16

16 But notice that in Estonian the verb motildetelda lsquoto thinkrsquo can govern the allative case


87

childish

lsquoI didnrsquot really think about money I was still so childish back thenrsquo (Blind 1992 23)

Finally it can be noted that al-postpositions occur in a number of fixed phrases and idioms

Even such cases can occasionally be translated with a Finnish l-case form such as North

Saami gozuid alde lsquoawakersquo = Finnish hereillauml The underlying nominative forms gohcu and

here do not occur as independent nouns in the languages (but cf North Saami gohcit lsquoto be

awakersquo and Finnish heraumltauml lsquoto awakenrsquo)

(31) Lean gozuid alde muhto buot orru dego niegus

be1SG STEMPLGA alde but all seem3SG like dreamLOC

daacutehpaacutehuvvame

happenPROG

lsquoOlen hereillauml mutta kaikki tuntuu tapahtuvan kuin unessarsquo

be1SG STEMPLADE but all feel3SG happenINF like dreamINE

lsquoI am awake but everything seems to be happening as if in a dreamrsquo (MAacute 1995)

It may be added that Finnic and Saami appear to have been close neighbors ever since their

divergence from a common protolanguage and as a result of millennia-long contacts Finnic

and Saami morphosyntaxes greatly resemble each other and the closest neighbors such as

Finnish and North Saami are rather isomorphic indeed One could hypothesize that this would

also have resulted in a gradual convergence of the functions of the Finnic l-cases and the

Saami postpositions However as will be shown below even the most remote members of the

two branches show significant similarities and this in turn is not fundamentally different from

the similarities with other geographically more remote descendants of the Uralic

postpositions

(i) Mina kuumlll ei motildetelnud rahale olin siis veel nii lapselik

1SG really NEG thinkCNGPST moneyALL bePST1SG then still so childish


88


From a qualitative perspective there is a clear correspondence between Saami al-postpositions

and Finnic l-cases especially in core local functions as showed in the previous subsection

Even more conclusive proof of their historical connection can be provided through a

quantitative analysis of the material According to our calculations as many as 1272ndash1321

out of the 1963 al-postpositions in our research material can be naturally translated into

Finnish with an l-case form this amounts to about two thirds of all tokens (65ndash67)

In order to see the frequency of various functions of the al-postpositions and to study

the correspondences with Finnish l-cases for each group of functions separately we have

divided the tokens into four broad semantic groups plus a residual group

a) Local expressions including metaphoric use of local expressions (see Examples 9ndash19

22ndash27)

b) Expressions of posture (see Examples 20ndash21)

c) Postpositions governed by various verbs and nouns (see Examples 28ndash30)

d) Fixed phrases and idioms (see Example 31) mdash Some other examples in the material

include maacutetkki alde [tripGA on] lsquowhile travelling on the journeyrsquo jurdagiid alde

[thoughtPLGA on] lsquolost in onersquos thoughtsrsquo olles mielaid alde [full mindPLGA on] lsquoin

onersquos right mindrsquo beassat niskki ala [getINF neckGA on] lsquoto get the upper handrsquo In this

group we have also included lsquofixed phrasesrsquo where the postpositional phrase has some

kind of idiomatic reading even though the complement of the postposition may freely

vary examples include Xa ala lsquoin addition to Xrsquo (eg buot dan ala lsquoin addition to all

thatrsquo) Xa ala lsquoafter completing Xrsquo (eg dien beaivaacutesa ala lsquoafter completing that dayrsquos

triprsquo)

e) Unclassified tokens mdash This residual group includes al-phrases with some kind of

unusual reading which nevertheless does not seem to be an established idiom or fixed

construction and also a couple of cases where the meaning of the phrase simply remains

unclear

The number of tokens in each category as well as their correspondences to Finnish l-cases is

shown in Table 5 below As the figures in the table reveal the great majority of tokens

involve expressions of locality (15621962 = 80) and the correspondence in this core group


89

is very strong as many as 71ndash74 of the tokens can be translated with l-case forms in

Finnish

alde ala ~ l-case alde ala ~ other

a) locality 1108ndash1149 (71ndash74) 414ndash455 (26ndash29)

b) posture 68 (94) 4 (6)

c) government 37ndash42 (23ndash26) 118ndash123 (74ndash77)

d) fixed phrase idiom 43ndash44 (41ndash42) 62ndash63 (58ndash59)

e) unclassified 16ndash18 (28ndash31) 40ndash42 (69ndash72)

Total 1272ndash1321 (65ndash67) 638ndash687 (33ndash35)

Table 5 The correspondences between North Saami al-postpositions and Finnish l-cases in different functional

domains

This kind of calculation naturally involves a certain degree of subjectivity Indeed the

differences between minimum and maximum percentages result from borderline cases where

it is hard to be sure whether the Finnish translation involving an l-case form is the most

natural one or where our native speakerrsquos judgments of naturalness differ However as such

unclear cases only amount to a few per cent of the material they do not have a significant

implication on the overall result ndash in the local functions the correspondence between Saami

al-postpositions and Finnish l-cases is pervasive

Moreover it is possible to conduct a more objective experiment by comparing texts that

have been translated from North Saami to Finnish or vice versa Our material includes four

such translated works of fiction From these we have also checked how often the North Saami

al-phrase matches an l-case form in the Finnish text ignoring our own intuition about

possible translations altogether the results can be seen in Table 4


90

Book l-cases al-postpositions

Tove Jansson Aacutehčči ja mearra (larr Muumipappa ja meri)17 144 170 (= 847)

Timo K Mukka Sipirjaacute (larr Laulu Sipirjan lapsista) 77 110 (= 70)

Jovnna-Aacutende Vest Čaacutehcegaacuteddaacutei nohkaacute boazobaacutelggis (rarr

Poropolku sammaloituu)

48 76 (= 632)

Annukka amp Samuli Aikio Girdinoaiddi baacuterdni (rarr

Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)

Table 4 The correspondence rates between North Saami al-postpositions and Finnish l-cases in translated texts

As shown in table 4 the objective test verifies our results in translated works nearly three

quarters of the North Saami al-postpositions correspond to an l-case form in the Finnish text

And one can add that this figure is still slightly lowered by discrepancies between the Saami

and Finnish texts In some cases the Finnish text does not contain an l-case ndash even though

such a translation would be perfectly possible ndash because the original text and the translation

do not exactly correspond to each other Compare the Saami example (32a) against its

equivalent in the Finnish translation (32b)

(32) a [ndash ndash] dat lea noidon mu oappa gollegoalsin

it be3SG conjurePSTPTCP 1SGGA sisterGA goldenmerganserESS

guhte ferte aacutevi al vuodjat

which must3SG openseaGA alde swimINF

lsquo[ndash ndash] she has conjured my sister into a golden merganser that must swim on the

open searsquo (Aikio amp Aikio 1978a 119)

b [ndash ndash] haumln on noitunut sisareni kultaiseksi

3SG be3SG conjurePSTPTCP sisterGEN1SG goldenTRANSL

koskeloksi niin ettauml haumlnen taumlytyy nyt uida meren

merganserTRANSL so COMP 3SGGEN must3SG now swimINF seaGEN

sylissauml

lapINE

17 This book seems to have been at least for the most part translated from the Finnish version Muumipappa ja

meri rather than from the Swedish original Pappan och havet


91

lsquo[ndash ndash] she has conjured my sister into a golden merganser so that she must now

swim on the bosom of the oceanrsquo (Aikio amp Aikio 1978b 119) (cf uida ulapalla

[swimINF openseaADE] lsquoswim on the open searsquo)

It is easy to compare the use of Finnish l-cases to North Saami in this manner as there are

plenty of texts that have been translated from one language to the other It is more difficult to

apply such a method to other Finnic and Saami languages but a comparison of North and

Lule Saami translations of the New Testament to the Finnish Olonetsian Estonian and

Livonian translations yields a rough picture of the correspondences In addition this makes it

possible to further match the material against the geographically and historically distantly

related Permic language Udmurt

As shown in Table 6 below the results of such a comparison are somewhat different A

major cause of the lower correspondence rates is the fact that source text of the translations of

the New Testament has usually been the Greek original in addition to which a variety of

different translations to other majority languages have been used in each translation process

On the other hand each translation may have its own theological bases so that the outcomes

are often not and have not even meant to be literal translations of the original text(s)

Language Bible translation Matches of matches

Lule Saami Aringdaring Testamennta (2000) 124 48

Udmurt Выль Сӥзён (1997) 174 67

Finnish Uusi testamentti (1992) 74 29

Olonetsian Uuzi Sana (2003) 137 53

Estonian Uus Testament (1997) 72 28

Livonian Ūž Testament (1942) 0 0

Table 6 The 258 tokens of North Saami al-postpositions in Ođđa Testamentta (the New Testament 1998)

matched with Lule Saami nal-postpositions (nanna nalta nali) Udmurt vil-postpositions (vilin viliś vile vilti

viliśen) and Finnic l-cases

At first glance the figures in Table 6 seem very odd It is unexpected that even between North

Saami and Lule Saami the correspondence rate is as low as 48 as these languages are so

closely related that they are even to a fair extent mutually intelligible Even more peculiarly

the correspondence rate between North Saami and Olonetsian is higher (53) than that

between North Saami and Lule Saami The correspondence rates with Finnish and Estonian


92

are much lower which is again surprising as Olonetsian is very closely related to Finnish In

Livonian no matches can be found but this is simply due to the fact that l-cases do not even

exist in this language as productive members of the case system Quite strangely the highest

correspondence rate occurs between the two most distantly related languages the match rate

of North Saami al-postpositions and their Udmurt cognates vil-postpositions is as high as

67 despite that these two languages are both geographically and taxonomically very far

from each other

One should note though that it is not the high rate of correspondence to Olonetsian and

to Udmurt which is surprising it was already shown that there is a very high rate of

correspondence between North Saami al-postpositions and Finnish l-cases in local functions

and comparing North Saami to Olonetsian one only expects the same result The high rate of

correspondence between North Saami and Udmurt also matches well with the previous

observation that the basic local functions of these postpositions were inherited from Proto-

Uralic already (see 31) Comparing these three translations to each other one finds as many

as 99 cases out of 258 (38) where the North Saami al-postposition is matched by both an

Olonetsian l-case and an Udmurt vil-postposition ndash in spite of three completely separate and

independent translation processes A great majority of these matches involve cases with a

concrete local function especially in the sense of lsquolocation on the upper surfacersquo This result

provides a good statistic confirmation for the Uralic origin of this function and reinforces the

view that the Finnic l-cases are grammaticalized reflexes of the Uralic uumll-postpositions

Compared to Olonetsian the significantly lower percentages of matches with Finnish

and Estonian result of certain idiosyncratic features of biblical language The dramatic

difference results from the fact that the phrase eatnama alde [earthGA alde] lsquoon earthrsquo has an

abnormally high frequency in the New Testament 49 cases out of 258 ie as many as 19

of all al-postpositional phrases As a correspondent of this phrase one finds an l-case form in

Olonetsian (mual) but a postpositional phrase in Finnish (maan paumlaumlllauml) and Estonian (maa

peal) This is because the corresponding l-case forms of maa lsquoearth landrsquo have become

lexicalized into a different meaning Finnish maalla Estonian maal lsquoin countrysidersquo

Incidentally the high frequency of eatnama alde is also responsible for a significant lowering

of the match rate between North Saami and Lule Saami as the Lule Saami translation

employs the local case form ednamin [earthINE] instead Compare the following example


93

(Luke 214)

(33) North Saami (OT)

a Gudni lehkos Ipmilii allagasas ja raacutefi eatnama alde

glory beIMP3SG GodILL placehighupLOC and peace earthGA alde

olbmuide geaid Ipmil aacuterpmiha

personPLILL whoPLGA God showmercy3SG

Lule Saami (AringT)

b Guddne Jubmelij allagisaacuten ja raacutefe suv gierugijda

glory GodILL placehighupINE and peace 3SGGEN lovedonePLILL

ednamin

earthINE

Udmurt (VS)

c laquoDan viliś Inmarli muzjem vilin kańillik aďamiosli ǯeč erikraquo

glory high GodDAT earth vilin peace personPLDAT good will

Finnish (Raamattu)

d Jumalan on kunnia korkeuksissa maan paumlaumlllauml rauha

GodGEN be3SG glory placehighupPLINE earthGEN paumlaumlllauml peace

ihmisillauml joita haumln rakastaa

personPLADE whoPLPTV 3SG love3SG

Olonetsian (US)

e Kunnivo Jumalale uumllimaumlzes taivahas i mual rauhus

glory GodALL highSUPINE heavenINE and earthADE peace

rahvahile kudamii Haumli suvaiččou

people(PL)ALL whoPLPTV 3SG love3SG

Estonian (Piibel)

f bdquoAu olgu Jumalale kotilderges ja maa peal rahu

glory beIMP3SG GodALL placehighupINE and earthGEN peal peace

inimestest hea meelrdquo

personPLELA good mood

Livonian (UT)

g Ouv volgotilde yļižis Jumalotilden ja mā pǟl arm

glory beIMP3SG placehighupINE GodDAT and earthGEN pǟl peace

rovvotilden jotildeva mēļ

peopleDAT good mood


94

lsquoGlory to God in the highest and on earth peace to men on whom his favor restsrsquo

If one removes the 49 tokens of eatnama alde from the material the match rates for Finnish

Karelian and Olonetsian become more uniform Still it is noteworthy that the percentages are

rather low compared to the translations of works of fiction listed in Table 5 In addition to the

general differences between biblical translation procedures this is also caused by another

peculiarity of biblical language namely the high frequency of al-postpositions that have a

complement with a human referent Outside the Bible these kinds of phrases are at all not

common in North Saami and the few that occur in the rest of our material pertain to more or

less unusual states of affairs cf njoarai godena olbmaacute nala lsquopoured moonshine on the manrsquo

in (29) However in the New Testament such cases are very common Mun bijan Vuoigŋan

su ala [ndash ndash] lsquoI will put my Spirit on himrsquo (Matthew 1218) [ndash ndash] seavdnjat gahčai

noidošeaddji ala [ndash ndash] lsquodarkness came over him [ldquoover the sorcererrdquo]rsquo (Acts 1311) [ndash ndash]

bohkaacuteid ja vuovssaacuteid varra ja guigguid gunat mat riškkuhuvvojit buhtismeahttumiid ala [ndash ndash

] lsquoThe blood of goats and bulls and the ashes of a heifer sprinkled on those who are

ceremonially uncleanrsquo (Hebrews 913) [ndash ndash] almmis gahčče olbmuid ala stuora

čuođibuddaacutesaš čuorbmasat lsquofrom the sky huge hailstones of about a hundred pounds each fell

upon menrsquo (Revelation 1621) Due to the fact that l-case forms of nouns with human

referents have possessive and dative functions these kinds of Saami al-phrases can naturally

not be translated with them (see the discussion on Example (29) above) There are as many as

53 al-phrases of this kind in the New Testament ie 205 of all tokens This can be

considered highly atypical use of North Saami because the rest of our material includes less

than a dozen comparable examples

In total the tokens involving either the phrase eatnama alde lsquoon earthrsquo or a noun with a

human referent cover as many as 102 cases out of 258 in the New Testament (ie 395) If

these tokens which strongly deviate from normal use of North Saami al-postpositions are left

out of the count the remaining tokens correspond relatively well to the use of Finnic l-cases

For instance the correspondence rate between Finnish and North Saami rises up to 47 (74

cases out of 156) Thus one can say that despite a few peculiarities resulting from the

idiosyncratic properties of biblical language the material from the New Testament verifies

the close correspondence between al-postpositions (and Udmurt vil-postpositions) and the

Finnic l-cases at least as regards Finnish Olonetsian and Estonian


95


As the Saami al-postpositions and Finnic l-cases have been shown to correspond well in both

form and function we already have strong evidence for equating them etymologically This

argument can be further strengthened by examining the cognates of Saami al-postpositions in

Permic and Samoyed languages As already shown in Examples (1ndash8) the basic local use of

corresponding Komi and Nenets postpositions is quite similar

Komi ki vilin ~ gieđa alde ~ kaumldellauml lsquoon the handrsquo (1) ju vile ~ joga ala ~ joelle lsquoon(to) the

riverrsquo (2) etc

Tundra Nenets ŋardegwendeg ńińa ~ juolgesuorpmaid(an) alde ~ varpaillani lsquoon my toesrsquo (7)

serdeg‿ńin ~ jieŋa ala ~ jaumlaumllle lsquoon(to) the icersquo (8) etc

As mentioned earlier the Finnic l-cases have often been considered diachronically related to

Permic l-cases Quite like in Finnic in the Permic languages there is a series of three cases

formed with a coaffix -l- followed by a primary local case suffix the genitive (Komi -len

Udmurt -len) the ablative (Komi -liś Udmurt -leś) and the dative (-li in both languages)

However the functions of these cases are primarily possessive and never local (see eg

Baker 1985 131ndash132 147 Bartens 2000 82ndash83 94ndash98 325 333ndash335) whereas in contrast

the possessive use of Finnic l-cases is clearly secondary As the primary local use of the

Finnic l-cases is paralleled in Permic by vil-postpositions instead it is much more natural to

assume that these two are historically connected

In the previous subsection it was shown that the use of Udmurt vil-postpositions

corresponds quite well to that of both Saami al-postpositions and Finnic l-cases Bartens

(1978 140ndash141 148ndash150 187ndash188) has paid attention to the functional similarity of Saami

al- and Permic vil-postpositions on a purely synchronic level Some quite prototypical

examples of such functions and their equivalents in Finnic can be seen in the following

(Matthew 1618)

(34) a Ja mun cealkkaacuten dutnje ahte don leat Biehtaacuter ja daacuten

and 1SG say1SG 2SGILL COMP 2SG be2SG Peter and thisGA

baacutevtti ala mun huksen girkon ja jaacutepmima riikka

rockGA ala 1SG build1SG churchGA1SG and deathGA kingdomGA

poarttat eai vuoitte dan (OT)

gatePL NEG3PL winCNG thatGA


96

b Ja maringn dunji javlav daringn le Petrus Baacutekte ja dan baacutektaacutej

and 1SG 2SGILL say1SG 2SG be2SG Peter rock and thatGEN rockILL

iehtjam girkkov tsieggiv man badjel

REFLGEN1SG churchACC raise1SG whatGEN over

jaacutebbmekaacutejmo uvsa e goassak faacutemov oattjo (AringT)

kingdomofthedeadGEN doorPL NEG3PL ever thatGA authorityACC getCNG

c Mon tinid veraśko ton ndash Petr ta iz vile Mon Asleśtim

1SG 2SGDAT say1SG 2SG Peter this rock vile 1SG REFLABL1SG

Čerkme kildito adlen kapkajez uz vormi

churchACC1SG foundFUT1SG hellGEN gateDEF NEGFUT3SG winCNG

soje (VS)

thatACC

d Ja minauml sanon sinulle Sinauml olet Pietari ja taumllle kalliolle

and 1SG say1SG 2SGALL 2SG be2SG Peter and thisALL rockALL

minauml rakennan kirkkoni Sitauml eivaumlt tuonelan

1SG build1SG churchGEN1SG thatPTV NEG3PL kingdomofthedeadGEN

portit voita (Raamattu)

gatePL winCNG

e I minauml sanon sinule sinauml olet Pedri Kallivo i taumllle

and 1SG say1SG 2SGALL 2SG be2SG Peter rock and thisALL

kallivole minauml puumlstuumltaumln oman uskojien kanzukunnan

rockALL 1SG raise1SG REFLGEN believerPLGEN nationGEN

Uadun vaumlgi ei voita sidauml (US)

hellGEN force NEG3SG winCNG itPTV

f Ja mina uumltlen sulle Sina oled Peetrus ja sellele kaljule

and 1SG say1SG 2SGALL 2SG be2SG Peter and thatALL rockALL

ma ehitan oma koguduse ja potildergu vaumlravad ei

1SG build1SG REFLGEN congregationGEN and hellGEN gatePL NEG

saa sellest votildeitu (Piibel)

getCNG thatELA victoryPTV

g Aga ma kītotildeb ka sinnotilden ku sa ūod Petrus ja sīe

but 1SG say1SG also 2SGDAT COMP 2SG be2SG Peter and thatGEN

kivmaumlg pǟl [] ma tieb ylzotilde entš lātkub ja

rockGEN pǟl 1SG make1SG up REFLGEN congregationGEN and


97

eļ vǟrotilded aumlb votildeit vindotilde tǟnda (UT)

hellGEN gatePL NEG get3PL winINF thatPTV

lsquoAnd I tell you that you are Peter and on this rock I will build my church and the

gates of Hades will not overcome itrsquo

(Mark 825)

(35) a Jesus bijai fas gieđaidis su čalmmiid ala

Jesus putPST3SG again handPLGA3SG 3SGGA eyePLGA ala

daacutel čielggai oaidnu ja olmmaacutei lei buoriduvvon ja

now clearPST3SG sight and man bePST3SG healPASSPSTPTCP and

oinnii buot čielgasit (OT)

seePST3SG all clearADV

b Jesus aacutejn nuppaacutedis giedajdis aringlmmaring tjalmij nali biejaj

Jesus again secondELA handPLACC3SG manGEN eyePLGEN nali putPST3SG

ja aringlmmaring tjalme dal tjielggin buorraacutenij ja gaacutejkka

and manGEN eyePL now clearPST3PL getwellPST3SG and allACC

tjielggasit vuojnnegaringdij (AringT)

clearADV seeINCHPST3SG

c Noš ik solen śin vilaz kize ponem no učkini

but DPT 3SGGEN eye vile3SG handACC3SG putPST23SG and lookINF

kosem So burmem no vańze čilkit adʒini

orderPST23SG 3SG behealedPST23SG and allDEFACC clear seeINF

kutskem (VS)

beginPST23SG

d Jeesus pani uudestaan kaumltensauml miehen silmille ja nyt

Jesus putPST3SG again handPL3SG manGEN eyePLALL and now

taumlmauml naumlki tarkasti (Raamattu)

this seePST3SG preciseADV

e Iisus uvvessah pani kaumlit miehen silmile i mies

Jesus again putPST3SG handPL manGEN eyePLALL and man

kačoi tarkazeh Haumli oli parandunnuh da

lookPST3SG precisely 3SG bePST3SG getwellPSTPTCP and

naumlgi kai selgiesti (US)



98

f Seejaumlrel pani Jeesus uuesti kaumled ta silmadele ja ta

thereupon putPST3SG Jesus again handPL 3SGGEN eyePLALL and 3SG

sai taumliesti terveks ja naumlgi kotildeike selgesti (Piibel)

getPST3SG fully healthyTRANSL and seePST3SG allPTV clearADV

g Siz ta tegiž paņ kaumldud taumlm sīlmad pǟlotilde [] ja se

then 3SG again putPST3SG handPL 3SGGEN eyePLGEN pǟlotilde and it

vaņtliz ja voļ tierrotildeks tiedotildet ja

lookPST3SG and bePST3SG healthyTRANSL makePASSPSTPTCP and

neiz ammotilde sieldistiz (UT)

seePST3SG allPTV clearADV

lsquoOnce more Jesus put his hands on the manrsquos eyes Then his eyes were opened his

sight was restored and he saw everything clearlyrsquo

(Acts 760)18

(36) a De son luoitaacutedii čippiidis ala ja čuorvvui alla

then 3SG descendPST3SG kneePLGA3SG ala and shoutPST3SG high

jienain [ndash ndash] (OT)

voiceCOM

b Buolvatjij nali luojttaacutedij ja jieddnaacutet tjuorvoj [ndash ndash] (AringT)

kneeDIMPLGEN nali descendPST3SG and loudADV shoutPST3SG

c Sobere piďes vilaz sultem no badʒim kuarajen

thereupon knee vile3SG settlePST3SG and big voiceINS

keśiśkisa veram [ndash ndash] (VS)

shoutCVB sayPST23SG

d Haumln vaipui polvilleen ja huusi kovalla aumlaumlnellauml

[ndash ndash] (Raamattu)

3SG descendPST3SG kneePLALL3SG and shoutPST3SG hardADE voiceADE

e Haumli pakui polvilleh da kirgai kovah [ndash ndash] (US)

3SG fallPST3SG kneePLALL3SG and shoutPST3SG hardILL

f Ja ta laskus potildelvili ning huumluumldis suure

and 3SG descendPST3SG ononersquosknees and shoutPST3SG bigGEN

haumlaumllega [ndash ndash] (Piibel)

18 The Estonian and Livonian adverbs potildelvili (34f) and puoļļindžotildel (34g) are explained in Section 34


99

voiceCOM

g Aga puoļļindžotildel eitotilden ōriz ta vegiz ȳoumllkotildeks

but ononersquosknees fallPSTPTCP shoutPST3SG 3SG forcefulGEN voiceCOM

[ndash ndash] (UT)

lsquoThen he fell on his knees and cried out [ndash ndash]rsquo

Besides the New Testament as another point of comparison one can use the material Reacutedei

(1962 11ndash35) presents in his monograph study of Komi postpositions Reacutedei cites a total of

169 usage examples of the postpositions vilin lsquoonrsquo viliś lsquooff fromrsquo ja vile lsquoontorsquo in various

local functions According to our calculations at least 96 (57) of these can be naturally

translated with an l-case form in Finnish In addition to local functions Reacutedeirsquos study also

includes numerous examples of postpositions in more marginal functions such as

postpositional phrases governed by various individual verbs The set of examples Reacutedei has

chosen for his study is naturally not statistically representative of the use of these

postpositions but even if one were to calculate all the cases listed the resulting

correspondence rate is 36 117 cases out of 321

Regrettably from the Samoyed languages even less suitable material is available for

comparison However Mikola (1975 45ndash50) gives a total of 79 examples of the use of the

Nenets postpositions ńińa lsquoonrsquo ńiddeg lsquooff fromrsquo and ńih lsquoontorsquo and this material already gives

a rough picture of their basic functions Roughly two fifths (at least 31) of the examples can

be naturally translated with a Finnic l-case form Even though such a limited material does

not give a statistically reliable picture of the use of Nenets ńi-postpositions it still

demonstrates that the basic local functions of the Uralic uumll-postpositions are quite similar not

only between Saami and Permic languages but also with Nenets As Saami Permic and

Samoyed (Nenets) are only extremely remotely related branches of Uralic which have not

been in any known areal contact with each other these functions can be quite reliably

reconstructed into Proto-Uralic19 Further it may be noted that Uralic uumll- has also survived

in Western Mari in which the functions of the postpositions βǝlnǝ lsquoonrsquo βǝ(l)kǝ lsquoontorsquo and

βǝlec lsquooffrsquo largely correspond to those of their Saami Permic and Samoyed equivalents

19 It is hardly necessary to mention that very recently certain dialects of Saami Komi and Nenets have come

into contact in Northern Russia and on the Kola Peninsula These recent contacts naturally cannot explain any

similarities between the use of old Uralic postpositions in these languages


100

discussed above (eg ləm βǝlnǝ lsquoon the snowrsquo i βǝlnǝ lsquoon the icersquo stoumll βǝ(l)kǝ lsquoonto the

tablersquo and təl βǝlec lsquooff the firersquo etc see also Moisio amp Saarinen 2008 sv βǝl-)

The reconstruction presented above provides a very strong argument for equating the

Finnic l-cases with Uralic uumll-postpositions As it is recognized that Proto-Uralic already had

the postpositions uumll-nauml lsquoonrsquo uumll-tauml lsquooff fromrsquo and uumlli-ŋ lsquoontorsquo and in Proto-Finnic one

finds the highly similar case suffixes -l-nA -l-tA and -l(l)-en in the same function it is

easy to believe that these suffixes are originally agglutinated postpositions


As strong functional arguments have now been presented in favor of the lsquouumll-theoryrsquo it is

necessary to examine the phonological and morphological aspects of the new explanation

The development of Finnic l-case endings out of Uralic uumll-postpositions is not

phonologically regular but the suffixation of independent postpositions cannot even in theory

be based on any sound law a regular development could only have resulted in uumll-

postpositions being retained as independent words However it is necessary to posit only

three irregular changes 1) univerbation of postpositional phrases by way of loss of the vowel

uuml 2) loss of the genitive ending -n in the adessive and the allative 3) adjustment to vowel

harmony The assumed development can be seen in Table 7

Pre-Finnic talja-n uumllnauml talja-n uumlltauml talja-n uumlli-ŋ

1) loss of uuml (taljanlnauml) (taljanltauml) taljanlen

2) loss of -n- taljalnauml taljaltauml taljanlen

3) vowel harmony taljalla taljalta taljallen

lsquoon the hidersquo lsquoofffrom the hidersquo lsquoonto the hidersquo

Table 7 The phonological development of uumll-postpositions into l-cases

In the scheme in table 7 one can also see other phonological developments namely the

assimilations ln gt ll (in the adessive) and nl gt ll (in the allative) the vowel lowering i gt

e in an unstressed syllable and the shift of the lative ending -ŋ into -n in word-final

position These can be interpreted as regular The change ln gt ll is well-established in

lexical items (eg Finnish halla lsquonight-frostrsquo lt šalna lt Proto-Baltic šalnā gt Lithuanian

šalnagrave) The change -ŋ gt -n has not traditionally been considered a sound law due to the

paucity of examples of a velar nasal in word-final position However the assumption of this


101

change makes it possible to equate the lative suffixes -n and -k and derive both of them

from the earlier form -ŋ the lative suffix attested in Mordvin and Samoyed languages

(Janhunen 1998 469 Bartens 1999 76 Ylikoski 2011 256ndash258) In Proto-Saami there was a

sound change -ŋ gt -k cf North Saami ala lt Proto-Saami ele-k lt Proto-Uralic uumlli-ŋ (cf

Sammallahti 1998 226)20

Regarding change 1) univerbation of the original postpositions and their complements

has led to the situation in which the original initial vowel of the postpositions has become

stressless and prone to loss One can add that vowel reduction and loss is especially common

in the case of close vowels such as uuml Unstressed close vowels become easily reduced and

lost apparently because their inherent phonetic duration is shorter than that of non-close

vowels (Laver 1994 435ndash436) This process can be seen in the phonological history of some

branches of Uralic as well Proto-Uralic unstressed i has become more frequently reduced or

lost in daughter branches than the open vowels a and auml eg in Proto-Mordvin (Bartens

1999 64ndash65) and Proto-Samoyed (Janhunen 1981 247ndash248 Sammallahti 1988 485) and

even in certain eastern dialects of Finnish (cf dialectal Finnish veś lsquowaterrsquo lt vesi but pesauml

lsquonestrsquo unchanged)21

Change 2) the loss of the genitive ending -n can be considered a direct consequence

of change 1) The loss of uuml would have resulted in the awkward consonant clusters nln and

nlt in the adessive and the allative and due to phonotactic restrictions such clusters would

have been simplified In fact it is doubtful whether forms such as taljanlnauml and taljanltauml

even occurred in the language at any period it would seem more natural to assume that the

genitive ending -n was lost at the same time with the vowel uuml Notably the earlier presence

of the genitive ending -n is revealed by the allative ending -llen the geminate lateral

20 Traditionally a large number of different Uralic directional case (lsquolativersquo) suffixes have been assumed at least

the ldquolativesrdquo -ŋ -n -ń -k -s and -j have been frequently reconstructed in studies on Uralic case systems

However it is not natural to assume that any real language would have had such a multitude of directional case

suffixes especially as no distinction between these suffixes has been established in terms of either their function

or their morphological distribution In our opinion it is much more plausible that the lsquolativersquo endings -k -n

and -ń (and perhaps also j) attested in various languages reflect an earlier -ŋ This question is however not

relevant to the origin of the Finnic l-cases our theory is not affected by whether the -n in the allative suffix -

llen reflects an earlier -ŋ or some other directional case suffix 21 In fact syncopes and apocopes seem to follow such a hierarchy that if non-close vowels are lost close vowels

must also be lost For instance in Estonian all final vowels were lost after long stressed syllables as in kaksi gt

kaks lsquotworsquo paksu gt paks lsquothickrsquo maksa gt maks lsquoliverrsquo But in contrast no Uralic language seems to have lost

non-close vowels in positions where close vowels have been preserved


102

reflects an earlier cluster nl which was assimilated in the same way as in compounds and on

word boundaries compare Finnish sellainen lsquothat kind ofrsquo taumlllainen lsquothis kind ofrsquo (ltlt sen

lajinen [itGEN kindof] taumln lajinen [thisGEN kindof]) and lttalon luonagt talol‿luona lsquoat the

housersquo In the context of the earlier lA-theory the geminate had been explained as a result of

influence of the adessive ending -llA but in our theory there is no need to resort to such an

explanation However it must be noted that the loss of the Proto-Finnic genitive ending -n in

constructions that would gradually develop into l-cases is conceptually independent of the

later sound changes that have lead to the variegated development and partly complete loss of

-n in individual Finnic languages

One should note that in the plural forms the phonological leap from postpositions to

case endings has been even smaller The genitive plural ending was originally merely -j as

still attested in Saami (Sammallahti 1998 70) the genitive plural endings -ten and -iten

attested in Finnic which combine one or two plural markers with the genitive singular ending

-n are later innovations Hence in the plural forms one only needs to postulate the loss of uuml

and an adjustment to vowel harmony eg talja-j uumll-nauml [hide-PLGEN on-LOC] gt talja-j-lnauml

gt Finnish taljoilla lsquoon hidesrsquo (note that the change aj gt oi is regular in Finnic see Kallio

2012a 2012b 234 Footnote 16) As -j- became interpreted as a plural marker such

cliticized forms as talja-j-lnauml have probably offered a strong analogical model for a singular

form talja-lnauml

At the stage when the uumll-postpositions had become phonologically reduced and

cliticized through changes 1) and 2) their eventual adjustment to vowel harmony was only

predictable One can also observe this in some later suffixation processes such as in the

obscured Finnish compounds taumllla(i)nen ~ taumlllauml(i)nen lsquothis kind ofrsquo (ltlt taumln lajinen) and

taumlmmouml(i)nen id (ltlt taumln moinen) The same has happened to case suffixes that originate from

postpositions in Hungarian another Uralic language with harmony eg the dative ending -

nak ~ -nek has become adjusted to vowel harmony but the original front vocalic form nek-

can still be seen in postpositional forms such as nek-em lsquoto mersquo nek-ed lsquoto yoursquo nek-i lsquoto

himherrsquo etc (on the etymology of the suffix see Kulonen 1993 85 Honti 2006)

Even though the development of uumll-postpositions into case endings is phonologically

quite a natural process one can still add that also the previously existing local case endings

have probably exerted an analogical influence As seen in Table 8 the primary Uralic local

case endings and especially the series of secondary s-cases have provided a model to which

the series of cliticized uumll-postpositions could be naturally adapted


103

Primary local cases s-cases uumll-postp gt l-cases

-nA -s-nA uumll-nauml gt -l-nA

-tA -s-tA uumll-tauml gt -l-tA

-ŋ ( ~ -n -k) -s-en ( lt -s-iŋ) uumlli-ŋ gt -ll-en

Table 8 The analogical influence of Uralic primary local cases and s-cases in the development of the l-cases

The phonological irregularities that must be assumed in the development of l-cases are rather

small and can be plausibly accounted for In fact one can note that the reductive

developments assumed here are rather minimal when compared against for instance the case

forms that have later developed out of postpositions in Veps (Tikka 1992) It can be added

that the earlier theory based on derivational suffix -lA was not entirely free of phonological

irregularities either If l-cases had developed out of the derivational suffix -lA one would

have to postulate an irregular loss of the vowel A before a primary case ending but the so-

called consonant-stems that developed through syncope in Uralic are regular only for Finnic

e-stems not for A-stems compare Finnish kieli lsquotonguersquo kiele-n GEN kiel-tauml PTV vs kala

lsquofishrsquo kala-n GEN kala-a PTV (lt kala-ta instead of kal-ta)

When estimating the plausibility of the development outlined above it is essential to

remember that the suffixation of independent postpositions is by definition an extraordinary

process which cannot be based on any regular phonological changes whatsoever A regular

development could only have led to the maintenance of the postpositions as independent

words While the Komi postpositional phrase mu vil-in [earth on-INE] lsquoon the earthrsquo can be

regarded as an expected reflex of its Uralic predecessor mixi-n uumll(i)-nauml the Olonetsian mua-l

[earth-ADE] as well as the Southern Permyak superessive form mu-vin id (lt mu vilin) to be

discussed in Section 36 below are from a purely phonological point of view anomalous

cognates of the Komi phrase

In addition to sound changes also one morphological change must be postulated If l-

cases indeed developed from postpositions the development probably had an intermediate

phase where a possessive suffix preceded the case ending instead of following it This is the

case for example with the comitative plural in Saami which developed from a postposition

guoimmi lsquowithrsquo In all other case forms possessive suffixes follow the case suffix in North

Saami but in the comitative plural the order is the reverse Compare North Saami maacutenaacute-i-

guin [child-PL-COM] lsquowith childrenrsquo vs maacutenaacute-id-an-guin [child-PL-1SG-COM] lsquowith my

childrenrsquo (lt maacutenaacute-id-an guoimmi [child-PLGEN-1SG with]) expected forms such as maacutenaacute-


104

i-guin-an [child-PL-COM-1SG] do not occur in the language at least yet Hence one must

assume that the development of the Finnic l-cases took place as shown in Table 9

lsquoon the backrsquo lsquoon hisher backrsquo

1 postpositional phrase selkauml-n uumll-nauml selkauml-n-sauml uumll-nauml

2 suffixation selkauml-lnauml selkauml-nsauml-lnauml

3 shift of suffix order selkauml-lnauml selkauml-lnauml-nsauml

4 Proto-Finnic selkauml-llauml selkauml-llauml-nsauml

Table 9 The morphological development of the l-cases

It is noteworthy that Livonian ndash where l-cases only occur as frozen relic morphemes ndash certain

adverbs seem to have preserved traces of the phase when the suffix order had not yet been

reverted eg sǟlganžotildel sǟlgandžotildel lsquoon onersquos backrsquo lt saumllkauml-nsauml-llauml ltlt saumllkauml-n-sauml uumll-nauml

[back-GEN-3SG on-LOC] and pȯļļindžotildel pȯļļizotildel lsquoon onersquos kneesrsquo lt polvi-nsa-lla ltlt polwi(-

j)-n-sa uumll-nauml [knee-(PL-)GEN-3SG on-LOC] Maumlgiste (1928) has tried to explain these forms

otherwise as possessive suffixes lost their productivity in Livonian Maumlgiste hypothesizes

that in such cases a possessive form would first have become lexicalized after which a case

ending would have been added to this lexicalized form This explanation is not convincing

however it is very hard to see why inflected forms such as saumllkaumlnsauml lsquohisher backrsquo and

polvinsa lsquohisher kneesrsquo would have become lexicalized in the first place and why adessive

forms would then have been formed from these lexicalized forms ndash only to become

lexicalized again in their turn It is also worth noting that Estonian where possessive suffixes

likewise lost their productivity has no such lexicalized possessive forms such as selgas

lsquohisher backrsquo or potildelves lsquohisher knee(s)rsquo or the like (potildelves is of course a regular inessive

singular form of potildelv lsquokneersquo) Instead in Estonian there is an adverb type selili ~ seljali lsquoon

onersquos backrsquo potildelvili lsquoon onersquos kneesrsquo etc (see [38] Section 33) Hence the Livonian adverbs

sǟlganžotildel and pȯļļindžotildel offer yet one more piece of evidence for the postpositional origin of

the l-cases


If the Finnic l-cases developed through agglutination of uumll-postpositions one expects that

these postpositions were lost as independent words at the same time For example the

Estonian comitative ending -ga and the Saami comitative plural ending -guin emerged when


105

an original postposition developed into a case ending (-ga lt kās -guin lt guoimmi) ndash they

have not been preserved as independent postpositions However in Finnish there are both l-

cases that developed from Uralic uumll-postpositions and ndash in a slightly different function ndash also

a series of postpositions and adverbs inherited from the same root yllauml lsquoabove on (of

clothes)rsquo yltauml lsquofrom above off (of clothes)rsquo and ylle lsquo(to) above over (putting) on (of

clothes)rsquo Below we will explain how this state of affairs can be accounted for

First it is worth noting that even though Finnish has a series of yl-postpositions their

use only extremely rarely corresponds to that of Saami al-postpositions In our entire material

there are only a handful of examples of this kind

(37) Seavdnjat seaivvui sullo ala [ndash ndash]

darkness landPST3SG islandGA ala

lsquoPimeys laskeutui saaren ylle [ndash ndash]rsquo

darkness descendPST3SG islandGEN ylle

lsquoDarkness came down over the island [ndash ndash]rsquo (Jansson 1990 144 1979 129)

The extreme rarity of these kinds of correspondences already suggests that the use of the

Finnish yl-series of postpositions in the lsquoaboversquo lsquooverrsquo function is in some way secondary

This is indeed obvious also from the fact such a function is not prominent in the Permic and

Nenets reflexes of the Uralic uumll-postpositions either

It is noteworthy that while modern literary Finnish has yl-postpositions in phrases such

as meren yllauml lsquoover the searsquo and saaren yllauml lsquoover the islandrsquo this is quite atypical of other

Finnic languages ndash and in fact also of the traditional Finnish dialects According to the data

in the Lexical Archive of Finnish Dialects the words yllauml yltauml and ylle are found mainly in

the western dialects in the eastern dialects they mostly occur just in idioms and fixed phrases

such as olla yllauml lsquoto be awakersquo and yltauml paumlaumlltauml XssA lsquocompletely altogether covered by

dirtied with Xrsquo Moreover even in the western dialects the words yllauml yltauml and ylle are

traditionally not used as postpositions but only as adverbs in reference to clothing eg takki

yllauml lsquowith a jacket onrsquo Examples of their use as postpositions are exceedingly rare in the

Lexical Archive of Finnish Dialects The following is apparently the only case which has a

noun (a participle functioning as a noun) as the complement of the postposition

(38) silkki levitettiiv‿vihittaumlvien ylle

silk spreadPSTPASS wedPASSPSTPLGEN ylle

lsquoA silk was spread over the bride and the bridegroomrsquo (LAFD Kankaanpaumlauml)


106

In addition one can find a couple of examples involving pronouns such as the following

(39) seoŋ‿kalarruumlsauml mun‿uumlllaumlin

itbe3SG fishGENtrap 1SGGEN yllauml1SG

lsquoIt is a fish trap [which I have] on mersquo (ie lsquoI am wearing it as if it were a piece of

clothingrsquo) (LAFD Lohja)

The situation is quite similar in other Finnic languages In Estonian the words uumlll lsquoonrsquo uumllt

lsquooffrsquo and uumllle lsquoon(to)rsquo are used in a similar way as adverbs in reference to clothing On the

other hand in Karelian Lude and Veps no cognates of these words are found at all ndash they

have been completely lost as adverbs as well

It is worth noting that even though the use of Finnish yllauml yltauml and ylle as postpositions

is extremely limited the prolative form of the same root yli ~ ylitse lsquooverrsquo is an entirely

common postposition This can be compared to the use of the morphologically fully

analogous postpositional series based on the root al- lsquounderrsquo Finnish alla lsquounderLOCrsquo alta

lsquounderABLrsquo alle lsquounderLATrsquo and ali ~ alitse lsquounderPROLrsquo All members of the latter series

frequently occur as postpositions Leino (1990 139) has paid attention to this discrepancy

between the two postpositional series He interprets the situation so that a new supplementary

series of postpositions is developing in Finnish paumlaumlllauml lsquoonLOCrsquo paumlaumlltauml lsquoonABLrsquo paumlaumllle

lsquoonLATrsquo yli ~ ylitse lsquoonPROLrsquo This supplementation can indeed be quite clearly seen by

comparing the relative frequencies of Finnish yl- paumlaumll- and al-postpositions with the

pronoun se lsquoitrsquo as their complement The numbers of tokens in Table 10 are based on searches

for the given character strings on the Google search engine on the World Wide Web

sen yllauml 909 sen paumlaumlllauml 28 700 sen alla 41 500

sen yltauml 28 sen paumlaumlltauml 1 230 sen alta 14 700

sen ylle 519 sen paumlaumllle 76 300 (cf) sen alle 40 500

sen yli 43 000 sen paumlaumllli ndash sen ali 271

sen ylitse 988 sen paumlaumlllitse 8 sen alitse 70

Table 10 The relative frequencies of Finnish yl- paumlaumll- and al-postpositions after the pronoun se lsquoitrsquo (Google

1142007 the search was limited to the top-level domain ldquofirdquo)


107

The statistics in Table 10 verify Leinorsquos main observation It must be pointed out however

that Leino does not even touch upon the possible reasons for the development of the

supplementary postpositional series paumlaumlllauml paumlaumlltauml paumlaumllle yli ~ ylitse Moreover contrary to

Leinorsquos claim this series is no longer ldquodevelopingrdquo even though yllauml yltauml and ylle have

limited use as postpositions in modern literary Finnish in old literary Finnish the

supplementation has been even more complete This can be seen in Table 11 where we

present the relative frequencies of yl- paumlaumll- and al-postpositions after words ending in the

nasal -n in the Corpus of Old Literary Finnish (Vanhan kirjasuomen korpus) the majority of

words ending in -n are genitive singular forms

-n yllauml 1 -n paumlaumlllauml 1 413 -n alla 2 066

-n yltauml 4 -n paumlaumlltauml 117 -n alta 92

-n ylle 1 -n paumlaumllle 4 491 (cf) -n alle 407

-n yli 145 -n paumlaumllli ndash -n ali ndash

-n ylitse 1 642 -n paumlaumlllitse 7 -n alitse 1

Table 11 The relative frequencies of yl- paumlaumll- and al-words after words ending in the nasal -n in old literary

Finnish The material derives from The Corpus of Old Literary Finnish (Vanhan kirjasuomen korpus Research

Institute for the Languages of Finland) containing approximately 3 200 000 words since 1543 until the early

1800s The orthographic variation in old literary Finnish has been normalized

In the entire material in the Corpus of Old Literary Finnish one can find only six instances

where yllauml yltauml or ylle is preceded by a word ending in -n and none of these instances is in

fact a postposition Instead all these tokens are adverbs that are coincidentally preceded by a

word-form ending in -n for example

(40) 2xi On tarpellinen ettauml otetan ylle paxummat waattet

secondly be3SG important COMP takePASS ylle thickCMPVPL garmentPL

eli pannan yllauml olewat waattet kiinni [ndash ndash]

or putPASS yllauml bePRSPTCPPL garmentPL closed

lsquoSecondly it is important that one puts on thicker clothes or buttons up the clothes one

has onrsquo (Suomenkieliset Tieto-Sanomat 171776)


108

On the other hand one can find some forms with possessive suffixes these kinds of cases

were already discussed above in connection with the use of yllauml yltauml and ylle in the Finnish

dialects

(1 Kings 1130)

(41) Ja Ahia rupeis sijhen uten hameseen cuin haumlnen yllaumlns

and A graspPST3SG itILL newILL dressILL which 3SGGEN yllauml3SG

oli ja rewaumlis cahdexitoistakymmenexi cappalexi [ndash ndash]

bePST3SG and tearPST3SG twelveTRANSL pieceTRANSL

lsquoAnd Ahijah took hold of the new cloak he was wearing and tore it into twelve piecesrsquo

(Biblia 1642)

(1 Samuel 175)

(42) Ja haumlnellauml oli waskilacki paumlaumlsaumlns ja suomuxen caltainen

and 3SGADE bePST3SG bronzehat headINE3SG and scaleGEN like

panzari yllaumlns [ndash ndash]

armor yllauml3SG

lsquoHe had a bronze helmet on his head and he wore a scale-like armorrsquo (Biblia 1642)

Thus as regards the use of the yl-series the Old Literary Finnish material thus yields quite

exactly the same picture as the dialect materials in the Lexical Archive of Finnish Dialects

the words yllauml yltauml and ylle have been primarily used as adverbs especially in reference to

clothing but extremely rarely as postpositions One can add that even the use of the Estonian

uumll- and peal-series conforms to this picture as seen in Table 12 (the very high frequency of

the phrases selle peale and selle uumlle is because these occur as fixed phrases with the meaning

lsquoin addition torsquo)

selle uumlll 6 selle peal 18 200 selle all 66 800

selle uumllt ndash selle pealt 15 600 selle alt 10 600

selle uumllle 17 selle peale 241 000 (cf) selle alla 26 500

selle uumlle 356 000 selle peali ndash selle ala 42 400

selle uumllitsiuumlletsi ndash selle pealitsi ndash selle

alitsialatsi ndash

Table 12 The relative frequences of Estonian uumll- peal- and al-words after the pronoun form selle lsquoitGENrsquo

(Google 1142007 the search was limited to the top-level domain ldquoeerdquo)


109

On the basis of the discussion above one can conclude that the use of Finnish yllauml yltauml and

ylle as postpositions is not a direct inheritance from Proto-Uralic Instead these words were

originally adverbs in Finnic and they have only become reintroduced as postpositions in

modern Finnish The innovation seems to be characteristic of literary language in particular

as these postpositions seem to have a rather formal tone and are less used in colloquial

speech Merimaa (2002 40ndash43) has pointed out that the words yllauml yltauml and ylle are described

as postpositions for the first time in Renvallrsquos grammar (1840) even though the prolative

postpositions yli and ylitse are mentioned in Finnish grammars since Petraeus (1649) already

It is also worth noting that both Renvall and the grammars from the late 19th century only cite

examples involving clothing eg Riisun takin yltaumlni [undress1SG coatGEN yltauml1SG] lsquoI take

my coat offrsquo

Thus leaving the innovations of modern literary Finnish aside the original Uralic uumll-

postpositions are attested in Finnic almost exclusively in prolative use22 The reason for this is

that the postpositions based on the root uumll- developed into case suffixes and only the

prolative form yli ~ ylitse was preserved as an independent postposition This raises an

obvious question why then did the prolative postposition not develop into a case suffix as

well The reason seems to be the analogical model provided by primary Uralic local cases

(locative -nA separative -tA and lative -ŋ) and especially the s-cases (inessive -s-nA

elative -s-tA and illative s-in) The tripartite structure of these case series provided a model

for the development for three lsquoexternalrsquo local cases but not for a lsquosuperprolativersquo case (see

Table 6 in Section 34)

The idea that uumll-words have been preserved when used as adverbs but changed into

case suffixes when used as postpositions can be compared to the emergence of the Estonian

comitative case In this case too the original postposition kaas developed into a case suffix

but was preserved as an independent adverb (see Figure 1)

22 The Livonian postposition irsquoļ has a much wider range in uses than its cognates elsewhere in Finnic because

the l-cases have not been preserved as productive case forms in this language (see eg Sjoumlgren amp Wiedemann

1861 37ndash38 72ndash74 Itkonen 1957a 310ndash311 Kettunen 1957 429ndash430 Itkonen 1957b 435ndash436 Halling 1996

1999)


110

postposition gt case adverb preserved

isaumln kaas isauml kaas

darr darr

Isaga isa ka

lsquowith fatherrsquo lsquofather alsorsquo

Figure 1 The development of the postposition and adverb kaas in Estonian

The same kind of result can also been seen in the Saami comitative plural (North Saami -i-

guin) which has developed from a postposition kuojmē(-n) cf North Saami guoibmi

lsquocompanion spousersquo (Korhonen 1981 225ndash226 Sammallahti 1998 69ndash70) At least in most

Saami languages the comitative plural ending is clearly a case suffix even though the South

Saami comitative plurals have also been analyzed as postpositional phrases (Bergsland 1946

148) Even in South Saami gujmie is clearly a marker of case regardless of whether it is

analyzed as a suffix or postposition because it is only used in connection of a plural form and

it is thus in complementary distribution with the comitative singular suffix -ine ~ -inie A very

rare exception to this pattern is that gujmie can also be attached to a phrase with plural

semantics but singular morphology eg aehtjie gon tjidtjien gujmie [father and motherGEN

gujmie] lsquowith father and motherrsquo (LS 19) -n is the genitive singular suffix

Regardless of how South Saami comitative plurals are analyzed it is quite evident that

in most Saami languages the comitative plurals are true case forms that originated from a

postpositional phrase The postpositional background can be seen for instance in conjunction

reduction (eg aacutehkaacute-id ja maacutenaacute-iguin [wife-PLGA and child-PLCOM] lsquowith wives and

childrenrsquo) and from the fact that possessive suffixes precede the case ending (eg maacutenaacute-id-

an-guin [child-PL(GA)-1SG-PLCOM] lsquowith my childrenrsquo) instead of following it as in other

case forms In older North Saami texts the ending is occasionally even spelt as a distinct word

and attached to singular forms


111

(43) [ndash ndash] moft Ibmel su oskaldasvuođaines su vuoiŋas ja

how God 3SGGA faithfulnessCOM3SG 3SGGA spiritGA3SG and

engelidis guim divcodaeligme bokte sin varjali

angelPLGA3SG guin takecareANGA by 3PLGA protectPST3SG

lokkamaeligttom vaddoin ja oasetesvuođain [ndash ndash]

countless difficultyPLLOC and misfortunePLLOC

lsquondash ndash how God protected them from countless difficulties and misfortunes by taking care

of them with his faithfulness with his Spirit and angelsrsquo (Muitalaeliggje 11873 3)

Even so the use of this word as a true postposition is exceedingly rare in modern Saami

languages if the South Saami comitative plurals are analyzed as case forms rather than

postpositional phrases But in South Saami gujmie is still used as an independent adverb in

the meaning lsquoalongrsquo as in baringetieh gujmie [comeIMP2SG along] lsquocome alongrsquo The original

postposition has thus developed into a case suffix but the adverb has been preserved exactly

as in the case of the Estonian ga-comitative

Using the Estonian and Saami comitatives as parallels the development of Uralic uumll-

postpositions and adverbs in Finnic can be assumed to have taken place as shown in Figure 2


talja-n uumll-nauml talja uumll-nauml

darr darr

taljalla talja yllauml

lsquoon a hidersquo lsquo(with) a hide onrsquo

Figure 2 The development of uumll-postpositions and adverbs in Finnic


In addition to all the arguments above the uumll-theory receives further support from parallels in

other branches of the Uralic family A particularly illuminating parallel is provided by the

case system in the Southern Permyak dialects of Komi The original Komi vil-series of

postpositions ndash ie the etymological cognates of Saami al-postpositions ndash has developed into

a set of case suffixes in Southern Permyak dialects (Batalova 1982 91ndash98 Baker 1985 66ndash


112

68 175ndash191) The agglutination process which is evidently fairly recent is illustrated in

Table 13

superessive -l(l)in ~ -v(v)in lt vilin

superlative -l(l)e ~ -v(v)e lt vile

sublative -l(l)iś ~ -l(l)iś lt viliś ~ viliś

~ -v(v)iś ~ -v(v)iś

perlative -l(l)eť ~ -v(v)eť lt vileť

superterminative -l(l)eʒ ~ -v(v)eʒ lt vileʒ

Table 13 The external local cases in the Southern Permyak dialects of Komi

Bartens (2000 79) even calls these Southern Permyak case forms lsquoexternal local casesrsquo and

this choice of words indeed describes well their striking functional similarity to the Finnic l-

cases The basic local use of the Southern Permyak external local cases is quite like that of

Finnish l-cases as shown by the following examples

(44) a gor-le [ltlt gor vile] kaj

oven-le [ oven onILL] goIMP2SG

lsquomene uunillersquo

goIMP2SG ovenALL

lsquoGo onto the ovenrsquo (Batalova 1982 94)

b šontiśni gor-lin [ltlt gor vilin]

warmoneselfINF oven-lin [ oven onINE]

lsquolaumlmmitellauml uunillarsquo

warmoneselfINF ovenADE

lsquowarm oneself on the ovenrsquo (ibid 96)

c gor-liś [ltlt gor viliś] oz lećći

oven-liś [ oven onELA] NEGFUT3SG descendCNG

lsquoei laskeudu uuniltarsquo

NEG3SG descendCNG ovenABL

lsquois not coming down from the ovenrsquo (ibid 96)


113

(45) kajnite ibbesle [ltlt ibbes vile]

goINFACC2SG fieldPLle [ fieldPL onILL]

lsquomennauml pelloillersquo

goINF fieldPLALL

lsquogo (on)to the fieldsrsquo (ibid 95)

(46) koklas [ltlt kok vilas] sulale

foot-las [ foot onINE3SG] stand3SG

lsquoseisoo jaloillaanrsquo

stand3SG footPLADE3SG

lsquostands on his feetrsquo (ibid 94)

(47) sulali prontlas [ltlt pront vilas]

standPST1SG front-las [ front onINE3SG]

lsquoseisoin rintamallarsquo

standPST1SG frontADE

lsquoI stood on the front line [in battle]rsquo (ibid 95)

In addition to Komi dialects the initial stages of such a development can be seen in Inari

Saami and in the Eastern Finnmark dialects of North Saami In these languages the al-

postpositions (North Saami alde and ala Inari Saami alne and oolacirc) are often pronounced

phonologically reduced and they tend to come cliticized to the preceding noun Consider the

following Inari Saami example

(48) [ndash ndash] jȧ nūut tot vaaldij tom stuorra keeđgi oalgg-ool

and so it takePST3SG itACC bigATTR rockACC shoulderGEN-oolacirc

jȧ kuodij tom stuorra geeđgi đoho njarggeij vuȧstȧ jȧ

and carryPST3SG itACC bigATTR rockACC there capePLGEN against and

đaeliglle đot vaaldij oalgg-aln tom geeđgi maeligddal [ndash ndash]

then it takePST3SG shoulderGEN-alne itACC rockACC away

lsquoAnd so he took that big rock on his shoulder and carried that big rock over there

towards the land points and then he took that rock off his shoulderrsquo (IK 27)


114

Similar cliticization of the postpositions alde and ala is also extremely common in the Eastern

Finnmark dialects of North Saami even though this is not commonly represented in literary

usage

(49) paumllk‿al ~ paumllk‿acircl (ltbaacutelgaacute aldegt ) lsquoon the pathrsquo

paumllk‿ala ~ paumllk‿acircla (ltbaacutelgaacute alagt ) lsquoonto the pathrsquo

The cliticization of these Saami postpositions is also discussed by Bartens (1978 191ndash195)

see also IW (sv ale-)23 One can still add that also in other Saami languages one finds

evidence for the proneness of uumll-postpositions to become cliticized In almost all western

Saami languages in an area reaching from South Saami to the Western Finnmark dialects of

North Saami the reflexes of uumll-postpositions show an initial nasal n- cf South Saami

nelnie nelhtie nille Lule Saami nanna nalta nali North Saami (western Finnmark) nalde

nala The nasal is originally the Proto-Saami genitive singular ending -n which was attached

to the complement of the postposition This shows that these postpositions have had a

tendency of becoming prosodically attached to the preceding nouns and offers yet one more

argument for the idea that a similar process of agglutination process began also in Pre-Proto-

Finnic


At this point when we have already presented many kinds of evidence for the uumll-theory it is

worthwhile to compare the new explanation against the previous lA-theory The traditional

explanation ndash and the assumptions implicit in it ndash are illustrated in Figure 3

23 According to Sammallahti (1977 239) similar cliticizations also occur in the Eastern Eanodat dialects which

belong to the Western Finnmark dialect group Sammallahti cites the postpositional phrases riepan-jok(aa)‿

alaa lsquoonto the river Riebanjohkarsquo and riepan-jok(aa)‿alte lsquoon the river Riebanjohkarsquo which he considers

phrasal loans from the adjacent Eastern Finnmark dialects spoken in Anaacuter


115

Proto-Uralic lumi-n uumll-nauml polwi-j uumll-nauml

lsquoon the snow on onersquos kneesrsquo

(inherent ON-function)

darr darr

Pre-Finnic Oslash lume-l(a)-na polv-i-l(a)-na

(no inherent ON-function)

darr

Finnish lumella polvilla(an)

cf South Saami

lopmen nelnie boelvi nelnie

Figure 3 The marking of the ON-function from Proto-Uralic to modern Finnish according to lA-theory

In comparison to the uumll-theory the major weakness of the lA-theory is that it presupposes a

much more complicated path of development The comparative method shows that Proto-

Uralic used a set of uumll-postpositions to mark the ON-function Therefore the lA-theory forces

one to assume that these postpositional phrases were replaced in Pre-Finnic with derivatives

with the suffix -lA- The uumll-postpositions with an inherent ON-function would have become

lost and at the same time the function would have been taken over by lA-derivatives ndash even

though such a function has never been attested in the derivational suffix itself Such a path of

development seems already in itself unlikely and it is made all the more improbable by the

fact that the Finnic l-cases and the Uralic uumll-postpositions show striking correspondence in

both form and function If one were to accept the lA-theory this correspondence would have

to be interpreted as an odd coincidence

Also typological arguments favor the uumll-theory One should note that local cases with an

ON-function are typologically quite rare usually location on the vertical axis is expressed with

adpositions but not with case endings (cf Blake 2001 151ndash154 Levinson 2003 98ndash110

Ojutkangas 2005 529ndash530) In addition to Finnic languages in the Uralic family only

Hungarian and the Southern Permyak dialects have these kinds of local case forms As

already mentioned the Southern Permyak case suffixes developed from postpositions and as

regards Hungarian at least the endings of the sublative and the delative also have a

postpositional background (Papp 1968 154 Kulonen 1993 84) The ending -en-on-oumln of

the superessive case may be an exception as it has been considered to derive directly from the

Proto-Uralic locative suffix -nA but even though this view is commonly accepted we must


116

point out that so far no one has presented an explanation to how the originally unmarked local

case might have developed a more limited and highly marked ON-function It may also be

noted that probably the closest functional equivalent to the Finnic l-cases in Indo-European is

the Ossetic adessive in -yl (Iron) ~ -baeligl (Digor) (see eg Thordarson 2009 153ndash154) For

example the adessive form zaeligxx-yl [earth-ADE] lsquoon the earthrsquo goes back to the Proto-Indo-

European words (s-)h1upeacuter(i) and dheacuteĝhōm (yielding eg Latin super humum id) and it is

therefore fully analogous to that of Finnic (eg Olonetsian mua-l id) and the newly emerged

superessive case in Southern Permyak (mu-vin id lt mu vilin) discussed in Section 36

above However we must conclude that the putative development of external local functions

from the derivational suffix -lA is backed by no well-attested functional parallels in the other

Uralic languages and we are not aware of such parallels in any other languages either

On the other hand there is at least one functional argument that could potentially

support the traditional lA-theory it is not inconceivable that an oikonym suffix could develop

into a local case marker considering the etymologies of French chez lsquoatrsquo and Mainland

Scandinavian hos id that go back to Latin casa lsquohousersquo and Scandinavian hus id

respectively Further it is not impossible that such locatives may later acquire possessive

functions (cf Section 42 below) As pointed out by Plank (2015 81) the locative form gehi

[houseLOC] of Pāli geha lsquohousersquo has developed ndash via locative functions ndash into the new

genitive case suffixes -gē and -ge in Sinhalese and Maldivian respectively (eg South

Maldivian goviyā-ge daruvō [farmer-GEN children] lsquothe children in the farmerrsquos [house]rsquo gt

lsquothe children of the farmerrsquo) However such unheard-of typological parallels to support the

received view on the origin of the Finnic l-cases do not alleviate the fact that the most original

function of the l-cases is evidently identical to that of postpositions formed from the Proto-

Uralic relational noun root uumll(i)- lsquoplace up or aboversquo and their descendants in a number of

modern Uralic languages It is highly improbable and without typological parallels that

oikonym derivatives in -lA would have initially superseded the Proto-Uralic uumll(i)-

postpositions in their concrete highly specialized yet universal functions ndash presumably also

supported by the formally and functionally analogous Proto-Finnic relational noun root al-

(lt Proto-Uralic il(a)-) lsquounderrsquo

From the perspective of both linguistic typology and the comparative method it is thus

quite natural to assume that the Finnic l-cases developed from independent postpositions This

theory is also in accordance with Occamrsquos Razor as one can postulate a much less

complicated path of development than is necessary in the lA-theory It is not necessary to


117

postulate any changes in the basic functions of the elements in Proto-Finnic but only in their

form postpositions have changed into case endings (see Figure 4)



darr

Pre-Finnic lume-l-na polv-i-l-na

darr


cf South Saami


Figure 4 The marking of the ON-function from Proto-Uralic to modern Finnish according to uumll-theory

In the same way several quite prototypical Finnish l-case forms can be explained as directly

inherited from Proto-Uralic postpositional phrases (see Figure 5)

Proto-Uralic

jaumlŋi-n uumll-nauml

lsquoon the icersquo

darr

kaumlti-n uumll-nauml

lsquoon onersquos handrsquo

darr

wolka-n uumll-nauml

lsquoon onersquos shoulderrsquo

darr

tuli-n uumll-nauml

lsquoon the firersquo

darr

Finnish jaumlaumlllauml kaumldellauml olalla tulella

cf North Saami jieŋa alde

~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)

Figure 5 Some Finnish l-case forms that are directly inherited from Proto-Uralic postpositional clauses


At this point it is useful to examine the origin of the non-local functions of Finnic l-cases We

will demonstrate that the development of possessive and instrumental functions in Finnic can

be quite naturally accounted for in framework of the uumll-theory Even so the arguments and

explanations presented in the following subsections have no real implication to our theory


118

regardless of what the actual origin origin of the l-cases is their possessive and instrumental

functions have in any case been practically unanimously considered secondary


Especially in the Northern Finnic languages the adessive is used in an instrumental function

In Saami al-phrases very rarely display similar semantics but instrumental uses are not

altogether unattested Nielsen (1979 sv acirclde) and Nickel (1994 168) mention the following

example which Nickel classifies as a metaphorical local phrase

(50) Daacuten biepmu alde ii eale gal guhkaacute

thisGA foodGA alde NEG3SG liveCNG indeed foralongtime

lsquoTaumlllauml ruoalla ei elauml kyllauml kauaarsquo

thisADE foodADE NEG3SG liveCNG indeed longtimePTV

lsquoOne wonrsquot survive long on this food for surersquo (Nickel 1994 168)

As pointed out by Ylikoski (2006 44ndash45) these kinds of alde-phrases can be used

interchangeably with comitative case forms which are the most common way to express

instrumentality in North Saami cf Daacuteinna biepmuin [thisCOM foodCOM] ii eale gal guhkaacute

One can also find other types of examples where the functions of an al-phrase and a

comitative form come close to each other eg a thing on which someone or something is

carried is usually simultaneously also an instrument for carrying and in such a context it

essentially irrelevant which form is used a postpositional phrase (51a) and a comitative form

(52a) are practically in a free variation with each other It is worth noting that in the

corresponding Lule Saami text the postpositional phrase (52b) and the comitative form (52b)

are used in exactly opposite to North Saami

(Luke 518)

(51) a Muhtun olbmaacutet gudde dohko laacutemis olbmaacute guoddinseaŋgga

some manPL carryPST3PL there lame manGA carryingbedGA

alde (OT)

alde

b De baringhtin soabmaacutesa guoddemlaacutetjujn gaacutellnam

then comePST3PL somePL carryingbedCOM beparalyzedPSTPTCP


119

aringlmmaringv guotte [ndash ndash] (AringT)

manACC carryCVB

c Paikalle tuli miehiauml jotka kantoivat vuoteella

placeALL comePST3SG manPLPTV whichPL carryPST3PL bedADE

halvaantunutta (Raamattu)

beparalyzedPSTPTCPPTV

lsquoSome men came carrying a paralytic on a stretcher [ndash ndash]rsquo

(Mark 655)

(52) a [ndash ndash] ja doapmaledje buot siidaguimmiid mielde ja

and hurryPST3PL all neighborPLGA with and

guoddigohte buhcciid guoddinseaŋggaiguin dohko gos

carryINCHPST3PL sickPLGA carryingbedPLCOM there where

gulle su leamen (OT)

hearPST3PL 3SGGA bePROG

b [ndash ndash] ja gaacutehtjadin aringbbaring baacutejke skihppij lusi ja

and hurryPST3PL whole placeGEN sickpersonPLGEN to and

de sijaacutejt guoddin laacutetjoj nanna daringhku garingnnaring

then 3PLACC carryPST3PL bedPLGEN nanna there where

gullin saringn lij (AringT)

hearPST3PL 3SG bePST3SG

c Sairaita alettiin kantaa vuoteillaan sinne missauml

sickPLPTV beginPSTPASS carryINF bedPLADE3PL there where

Jeesuksen kuultiin olevan (Raamattu)

JesusGEN hearPSTPASS beINF

lsquoThey ran throughout that whole region and carried the sick on stretchers to

wherever they heard he [Jesus] wasrsquo

While North Saami guoddinseaŋgga alde lsquoon a stretcherrsquo (51a) is literally a local adverbial

phrase and guoddinseaŋggaiguin lsquowith stretchersrsquo (52a) in turn an instrumental adverbial

phrase it is neither possible nor even necessary to determine whether vuoteella and vuoteilla

in the corresponding Finnish passages have a local or an instrumental function or even both


120

As already seen in (50) alde-phrases occasionally occur also as more clearly non-local

instrumental function In our research material this can be seen in certain fixes phrases and

idioms such as the following (for more detailed discussion see Ylikoski 2006 44ndash45)

(53) Ieš-Pieti čuovvolii aacuterrat leastadialaš oskku ja šattai

Ieš-Pieti beginfollowingPST3SG early Laestadian faithGA and becomePST3SG

dovddus saacuterdnideaddjin guhte iežas burssa nalde finai

famous preacherESS which REFLGA3SG walletGA alde goPST3SG

saacuterdnemaacutetkkiin Suomas ja Norggas

preachingtourPLLOC FinlandLOC and NorwayLOC

lsquoIes-Pieti rupesi varhain seuraamaan lestadiolaista uskoa ja

Ies-Pieti beginPST3SG early followINF LaestadianPTV faithPTV and

haumlnestauml tuli tunnettu saarnaaja joka omalla kukkarollaan

3SGELA comePST3SG famous preacher who ownADE purseADE3SG

kaumlvi saarnamatkoilla Suomessa ja Norjassarsquo

goPST3SG preachingtourPLADE FinlandINE and NorwayINE

lsquoIeš-Pieti converted to Laestadianism at an early stage and he became a famous

preacher who made preaching tours to Finland and Norway at his own cost (ldquoon his

own purserdquo)rsquo (Kristiansen 2004b 39)

More straightforward correspondents to the instrumental use of the adessive can be found in

other Uralic languages viz in Mordvin The Uralic uumll-postpositions have not been preserved

in Mordvin in their original local functions they have been replaced with new postpositions

formed from a relational noun root lang- which is of obscure origin (Saarinen 2005)

Nevertheless the original Uralic separative form uumll-tauml is reflected in the Mordvin

postposition veľďe (Erzya) veľďauml (Moksha) which has a primarily instrumental function

The following examples which derive from Paasonenrsquos Mordwinisches woumlrterbuch (MW sv

veľďe) show that the function of veľďe is in many ways similar to the instrumental adessives

(the examples have been converted into a phonological transcription)

(54) piľgesur veľďe jakams (Cf (7))

toe veľďe goINF

lsquokulkea varpaillaanrsquo

goINF toePLADE3SG

lsquowalk on onersquos toesrsquo


121

(55) mon piks veľďe valgiń

1SG rope veľďe descendPST1SG

lsquolaskeuduin koumlydellauml (koumlyttauml pitkin)rsquo

descendPST1SG ropeADE (ropePTV along)rsquo

lsquoI descended along a ropersquo

(56) meľ veľďe

mind velde

lsquomielellaumlaumlnrsquo

mindADE3SG

lsquogladly with pleasurersquo

(57) mon ramavtija sonze veľďe

1SG buyCAUS1SGgt3SG 3SGGEN veľďe

lsquoostatin sen haumlnellaumlrsquo

buyCAUSPST1SG itGEN 3SGADE

lsquoI made him buy itrsquo

Example (57) is especially remarkable as it employs veľďe as an agent marker in connection

with a causative verb Also the Finnish adessive case has developed the same function In

general the Mordvin examples listed above can be compared to Leinorsquos (1989 211) entirely

synchronic description of the use of the adessive case in Finnish ldquoEi ole vaikea konstruoida

esimerkkisarjaa puhtaasti spatiaalisesta adessiivin kaumlytoumlstauml instrumentaalisen ja jopa toisen

asteen agenttia osoittavaan [ndash ndash]rdquo (lsquoIt is not difficult to construct a series of examples from a

purely local use of the adessive to the instrumental and even to one expressing a second-level

agentrsquo) As an example of such a continuum he gives the following set of sentences

(58) a Pekka kuljetti lautalla Paavon saaresta

Pekka transportPST3SG raftADE PaavoGEN islandELA

lsquoPekka took Paavo off the island on a raftrsquo

b Pekka kuljetti veneellauml Paavon saaresta

Pekka transportPST3SG boatADE PaavoGEN islandELA

lsquoPekka took Paavo off the island onwith a boatrsquo


122

c Pekka ajoi reellauml tukit metsaumlstauml

Pekka drivePST3SG sledADE timberPL forestELA

lsquoPekka transported the timbers out of the woods onwith a sledrsquo

d Pekka ajoi hevosella tukit metsaumlstauml

Pekka drivePST3SG horseADE timberPL forestELA

lsquoPekka transported the timbers out of the woods with a horsersquo

e Pekka ajatti Paavolla tukit metsaumlstauml

Pekka driveCAUSPST3SG PaavoADE timberPL forestELA

lsquoPekka made Paavo transport the timbers out of the woodsrsquo (cf (57) ramavtija

sonze veľďe)

The usage of the Mordvin postposition veľďe corresponds quite well to that of the Finnish

adessive Thus the Mordvin examples offer good parallels for the development of

instrumental and even agent functions in the Finnish adessive case One should mention

though that there is a slight morphological discrepancy Mordvin veľďe reflects the Uralic

ablative case (uumll-tauml) and not the locative case (uumll-nauml) like the Finnish adessive This

distinction is not too great though as also the ablative case has limited instrumental use in

Finnic languages cf eg dialectal Finnish vaumlkiseltaumlaumln lsquoby forcersquo (vaumlki lsquocrowd strengthrsquo) and

Estonian vaevalt lsquowith difficultyrsquo (vaev lsquodifficultyrsquo) On the other hand one could also

surmise that the Mordvin form veľďe has some kind of irregular background for example the

Uralic il- lsquounderrsquo word family has given in Mordvin ndash in addition to the postpositions alo

lsquounderLOCrsquo aldo lsquounderABLrsquo and alov ~ aloŋ lsquounderLATrsquo ndash the derivative aldoń lsquolocated

under below [adjective]rsquo (Niemi amp Mosin 1995 sv) even though the expected form would

be aloń instead24


Possessive functions are one of the core functions of the l-cases in Finnic As noted above in

321 l-cases are not used in a local function with nouns or pronouns with human referents

because in such cases their use is restricted to possessive functions The complementary

distribution of local and possessive functions can be illustrated with the following examples

24 Note that there is a homonymous aloń which is both a genitive form and an adjective derivative of the word al

lsquoeggrsquo


123

(59) a Kirja on poumlydaumlllauml

book be3SG tableADE

lsquoThe book is on the tablersquo

b Otin kirjan poumlydaumlltauml

takePST1SG bookGEN tableABL

lsquoI took the book off the tablersquo

c Panin kirjan poumlydaumllle

putPST1SG bookGEN tableALL

lsquoI put the book on the tablersquo

(60) a Minulla on kirja

1SGADE be3SG book

lsquoI have a bookrsquo

b Ota kirja minulta

takeIMP2SG book 1SGABL

lsquoTake the book from mersquo

c Anna kirja minulle

giveIMP2SG book 1SGALL

lsquoGive the book to mersquo

Even though the al-postpositions in Saami are generally not used in possessive functions

there are nevertheless borderline cases that give some idea as to how the possessive functions

might have developed in Finnic Especially the postposition ala is sometimes used in a dative-

like function as in the following examples

(61) Dat mainna in leat duhtavaš lea ahte eanaš

it whatCOM NEG1SG beINF satisfied be3SG COMP most

ovddasvaacutestaacutedus gahččaacute moatti olbmo ala Kaacuteraacutešjogas

responsibility fall3SG fewGA personGA ala KaacuteraacutešjohkaLOC

lsquoSe mihin en ole tyytyvaumlinen on ettauml enin vastuu

it whatILL NEG1SG beCNG satisfied be3SG COMP most responsibility

lankeaa muutamalle ihmiselle Kaarasjoellarsquo

fall3SG fewALL personALL KaacuteraacutešjohkaADE


124

lsquoWhat Irsquom not satisfied with is that most of the responsibility falls on a couple of

people in Kaacuteraacutešjohkarsquo (MAacute 1995)

(62) [ndash ndash] Mathis M Sara fas oaivvildii staacutehta bidjat olu barggu

Mathis M Sara inturn meanPST3SG state putINF much workGA

orohagaid ala

herdingdistrictPLGA ala

lsquoMathis M Sara taas oli sitauml mieltauml ettauml valtio laittaa

Mathis M Sara inturn bePST3SG itPTV mindPTV COMP state put3SG

paljon tyoumltauml paliskunnillersquo

much workPTV herdingdistrictPLALL

lsquoMathis M Sara in turn was in the opinion that the state puts much work on the

reindeer herding districtsrsquo (MAacute 1995)

(Acts 126)

(63) Sii vuorbaacutededje dan guoktaacutesa gaskkas ja vuorbi gahčai

they castlotsPST3PL itGA twopeopleGA between and lot fall3SGPST

Mattiasa ala (OT)25

MatthiasGA ala

lsquoSen jaumllkeen he heittivaumlt miehistauml arpaa ja arpa lankesi

itGEN after 3PL throwPST3PL manPLELA lotPTV and lot fallPST3SG

Mattiaksellersquo (Raamattu)

MattiasALL

lsquoThen they cast lots and the lot fell to Matthiasrsquo

25 One can note that in the Greek original of the New Testament this passage contains the primarily local

preposition epi which has also been translated into Udmurt as vile

(ii) [ndash ndash] και επεσεν ο κληρος επι Μαθθιαν [ndash ndash] (NTGr)

and fallAOR3SG DEFSGM lot on MatthiasMACC

(iii) Pussi Matfij vile uśem (VS)

lot Matthias onILL fallPST23SG


125

It is also worth noting that in many Saami languages the reflexive pronoun shows a

supplementary paradigm where the local case forms diachronically reflect possessive forms

of the words alde and ala This is the case in North Saami as well as can be seen from the

following partial paradigm of the reflexive pronoun ieš

NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces

In the paradigm above the locative forms are diachronically nothing other than the

postposition alde lsquoonrsquo combined with possessive suffixes hence the use of a form such as

alddaacuten REFLLOC1SG has developed from the sense of lsquoon mersquo The background of the illative

forms is morphologically somewhat more complex a form such as alccen derives through an

irregular phonological development form earlier alla-sa-n with the same postpositional root

but a secondary possessive illative suffix -sa- preceding the possessive suffix In the dialects

one even finds forms such as alcce-sa-n with yet another secondary illative suffix added

Semantically though the background of the illative forms is wholly analogous to the locative

forms alccen lsquoto myselfrsquo developed its current function from an original meaning lsquoonto mersquo

The local case forms of the reflexive pronoun are used precisely in possessive functions

as the following examples reveal

(64) Ja juos alddiineaset ii leat ruhta de staacutehtta=han gal

and if REFLLOC3PL NEG3SG beCNG money then state=forsure indeed

sidjiide addaacute

3PLILL give3SG

lsquoJa jos heillauml itsellaumlaumln ei ole rahaa niin

and if 3PLADE REFLADE3PL NEG3SG beCNG moneyPTV then


126

valtio=han kyllauml heille antaarsquo

state=forsure indeed 3PLALL give3SG

lsquoAnd if they have no money themselves the state will give them for surersquo (Marastat

1991 19)

(John 7 17)

(65) Dat guhte daacutehttu dahkat su daacutehtu oažžu dovdat lea=go

3SG who want3SG doINF 3SGGA willGA get3SG feelINF be3SG=Q

oahppu Ipmilis vai mus alddaacuten

teaching GodLOC or 1SGLOC REFLLOC1SG

lsquoHaumln joka tahtoo tehdauml haumlnen tahtonsa saa tuntea on=ko

3SG who want3SG doINF 3SGGEN willGEN3SG get3SG feelINF be3SG=Q

opetus Jumalasta vai minulta itseltaumlnirsquo

teaching GodELA or 1SGABL REFLABL1SG

lsquoIf anyone wants to do Godrsquos will he will get to know whether the teaching is from God

or from me myselfrsquo (OT)






The following dialectal example involving the reflexive pronoun documented by Friis (1856

69) comes especially close to the possessive use of the l-cases in Finnic the phrase ješ aldam

consists diachronically of the same morphemes as the Finnish l-case form itsellaumlni

(66) ješ aldam laelig girje

REFL REFLLOC1SG be3SG book

lsquoMinulla itsellaumlni on kirjarsquo

1SGADE REFLADE1SG be3SG book

lsquoI myself have a bookrsquo (Friis 1856 69 lsquojeg selv har en Bogrsquo)


127

The attested ndash albeit very limited ndash possessive use of the Saami al-postpositions shows that it

is by no means unnatural that the grammaticalized uumll-postpositions have developed possessive

functions in Finnic In addition one can note that the development of local functions to

possessive ones is cross-linguistically quite common For instance the Russian preposition u

lsquoatrsquo is also used in possessive constructions eg u menja jesť kniga [at 1SGGEN be3SG book]

lsquoI have a bookrsquo (ldquothere is a book at merdquo) In most Saami languages possession is indicated

with the primarily local inessive or locative case eg North Saami mus lea girji [1SGLOC

be3SG book] lsquoI have a bookrsquo The starting point of such development can be seen in eg the

Siberian language Kolyma Yukaghir where the locative case suffix -ge is sometimes used

metaphorically in constructions resembling grammatical possession even though usually

possession is indicated in other ways (Maslova 2003 107 447ndash448)

(67) šoromo-ge qojl ninge-j

man-LOC god many-INTR3SG

lsquoMan has many godsrsquo (Maslova 2003 107)

In this connection we can briefly return to the possessive l-cases in Mari and Permic

languages which were already discussed earlier Mari has a dative case with the suffix -lan

(in West Mari -lan -laumln) and the easternmost dialects of the language also have an ablative

case with the suffix -leč (Alhoniemi 1985 44 52ndash54 61ndash62) The Permic languages in turn

have developed a series of three possessive cases consisting morphologically of the coaffix -

l- followed by a primary local case suffix the genitive (Komi -len Udmurt -len) the ablative

(Komi -liś Udmurt -leś) and the dative (-li in both languages) In the framework of the lA-

theory the Finnic Mari and Permic l-cases have been seen the result of convergent

development but in all languages the cases would ultimately have their origin in the ldquolocalrdquo

derivational suffix -lA The uumll-theory however implies that the Finnic and the Mari-Permic

l-cases cannot have a common background the Mari and Permic l-cases could not have

developed from uumll-postpositions because these postpositions have been retained as

independent words in these languages Moreover such an idea would also involve major

semantic difficulties as the Mari-Permic l-cases are almost exclusively possessive and they

do not show any trace whatsoever of an earlier ON-function that is inherent in uumll-postpositions

and the Finnic l-cases

Sometimes it has even been surmised that the possessive function of the Finnic l-cases

would be primary and they could thus be historically connected with the Mari-Permic l-cases


128

Anttila and Uotila (1984 127) maintain that possessive use could have developed via

reanalysis of oikonym derivatives based on the suffix -lA as follows setauml-lauml-nauml on peltoja

lsquounclersquos house has fieldsrsquo gtgt Finnish sedaumlllauml on peltoja lsquouncle has fieldsrsquo (cf Finnish setaumllauml

lsquounclersquos housersquo larr setauml lsquounclersquo) This suggestion is already made highly unlikely by the fact

that the oikonym derivatives in -lA have a marginal status and low frequency in the language

and if such a path of development is assumed it becomes very difficult to understand how the

concrete function of lsquolocation on the upper surfacersquo could have developed from much more

abstract possessive use Both historically and typologically it is more natural to assume that

the primary function of Finnic l-cases is local and that the possessive functions have

developed from metaphoric use of this local function and not from the reanalysis

constructions involving oikonym derivatives

In fact the uumll-theory reveals that the previous idea of a connection between Finnic and

Mari-Permic l-cases ndash either as cognate forms or as the result of convergent development of

the derivational suffix -lA ndash is based on circular reasoning This becomes evident from the

arguments that Bartens has presented in support of the equation of the Finnic and Permic l-

case forms (cf (59ndash60))

Ulkopaikallisuuden ilmoittaminen (esim kirja on poumlydaumlllauml panin kirjan poumlydaumllle otin kirjan poumlydaumlltauml) ei

kuitenkaan ilmeisesti ole ainakaan vanhempi funktio kuin itaumlmerensuomen l-sijojen habitiiviset ja datiiviset

funktiot (minulla on kirja anna kirja minulle ota kirja minulta) Sukukielissauml nimittaumlin ulkopaikallisuus

ilmaistaan tyypillisesti postpositiorakenteilla (esim komi kńigays pyzan vylyn rsquokirja on poumlydaumlllaumlrsquo pukti

kńigasouml pyzan vylouml rsquopanin kirjan poumlydaumlllersquo bośti kńigasouml pyzan vylyś rsquootin kirjan poumlydaumlltaumlrsquo) ja on

mahdollista olettaa ettauml alkuperaumlistauml on juuri postpositiorakenteiden kaumlyttouml taumlssauml funktiossa (Bartens 2000

83)

lsquoThe expression of external locality (eg kirja on poumlydaumlllauml panin kirjan poumlydaumllle otin kirjan poumlydaumlltauml) is

nevertheless apparently not at least an older function than the possessive and dative functions of the l-cases

(minulla on kirja anna kirja minulle ota kirja minulta) [cf (59ndash60)] In related languages external locality is

typically expressed with pospositional constructions (eg Komi kńigays pyzan vylyn lsquothe book is on the

tablersquo pukti kńigasouml pyzan vylouml lsquoI put the book on the tablersquo bośti kńigasouml pyzan vylyś lsquoI took the book off

the tablersquo) and it is possible to assume that exactly the use of postpositions is original in this functionrsquo

As one considers the issue from the perspective of the uumll-theory the latter sentence in the

quote above contradicts the first one Bartens is quite right in noting that the use of the Permic

vil-postpositions (and their cognates) to express external locality represents the historically

primary construction type Nevertheless even internal reconstruction confirms that the


129

expression of external locality is the primary function of the Finnic l-cases (see 31) Bartens

denies this but apparently only because she considers it to contradict the evidence from

Permic and other related languages But there is no contradiction whatsoever when the Permic

vil-postpositions are properly analyzed as cognates of the Finnic l-case endings

Thus there remain no valid arguments for connecting the primarily local l-cases in

Finnic with the possessive l-cases in Permic Indeed such an equation is methodologically

dubious in the first place the compared morphemes possess merely one matching

phonological segment (the consonant -l-) and a one similar function (possessive use) which

can be quite clearly shown as secondary in Finnic It becomes dangerously easy to find

accidental matches for grammatical morphemes of the shape -C- if cognates are sought over

a broad semantic spectrum For example in the Kolyma Yukaghir language (already

mentioned above in Section 42) there is an instrumental case with the suffix -le (Maslova

2003 77ndash78 104ndash105) and it would be methodologically wholly analogous to connect this

with the Finnic l-cases there is one identical segment (the phoneme -l-) and one similar

function (instrumental) which can be shown to be secondary in Finnic One should note that

chance resemblances of this kind can as easily come up between languages that are

genetically related not only between languages belonging to separate families like Finnish

and Yukaghir26

The origin of the Mari and Permic l-cases remains unexplained though equating them

with the derivational suffix -lA is not based on any more solid evidence that of the Finnic l-

cases as pointed out by Serebrennikov (1962 1963) Even though the question cannot be

scrutinized in detail here we can suggest a new hypothesis As mentioned above the

development of possessive functions from earlier local functions is typologically natural and

hence one could surmise that also the Mari-Permic l-cases may derive from some kind of

postpositions with local functions A candidate for such a source would be the postpositional

root reflected in North Saami lu- Finnish luo- lsquoatrsquo cf Finnish luona lsquoat in the vicinity ofrsquo

luota lsquofrom (the vicinity of)rsquo luo ~ luokse lsquoto (the vicinity of)rsquo No cognates for this root are

known outside Finnic and Saami but it is not at all impossible that its cognate is hiding in an

agglutinated form in the Mari and Permic l-cases ndash it is in fact necessary to assume that if

26 It is sometimes maintained that Yukaghir languages are related to Uralic (eg Nyikolajeva 2000 92ndash102 cf

also Nikolaeva 2006 viii amp passim) but no plausible arguments for this view have ever been presented (see

Aikio 2014 for discussion) and even if one subscribed to the idea of a Uralic-Yukaghir affinity there would of

course be no reason to assume a historical connection between the Finnic l-cases and the Yukaghir instrumental

case


130

these cases reflect earlier postpositions the original postpositions underlying them have not

been retained as independent words (cf 35) In other words the Mari and Permic l-cases

cannot derive from the Uralic uumll-postpositions as these postpositions were retained as

independent words in these languages

As regards the semantics of the Mari-Permic l-cases their development could be

compared eg to the Russian postposition u which was discussed earlier A particularly

illuminating point of comparison is offered by the lsquoatrsquo-series of local case in Veps which

developed through the agglutination of postpositions based on the root lo- the cognate of

Finnish luo- The semantics of these cases is predominantly local but sometimes their usage

comes close to possessive functions

(68) kaži goľu minu-лon meiďe-лon

cat always 1SGAPPR1 1PLAPPR1

lsquoThe cat is always at me at usrsquo (Kettunen 1943 369)

As also the Mordvin and Hungarian dative cases have similar postpositional backgrounds (see

Bartens 1999 79 Honti 2006 Ylikoski 2011) the development of local postpositions to

possessive case endings seems to be relatively common process in the Uralic languages

Nevertheless the explanation proposed for the Mari and Permic l-cases above is at this point

naturally still a mere hypothesis which requires more thorough scrutiny


Even though the uumll-theory offers a convincing explanation of the origin of the l-cases the

earlier lA-theory nevertheless includes some findings that can be incorporated in our new

model In addition to Finnic many other Uralic languages too possess derivatives based on a

semantically indeterminate local suffix -lA or -l(V) These kinds of derivatives are usually

formed from either a relational noun root or a monosyllabic pronoun root followed by a

coaffix In this way combinations of the suffix -lA and (local) case endings are used to form

various adverbs as in the following North Saami examples


131

a) pronoun root + coaffix -mpA- + suffix -lA- + local case ending eg daacutebbelis lsquocloser to

over herersquo (lt Proto-Saami tā-mpē-lē-snē [this-mpA-lA-INE]) daacute-ppi-l lsquofrom this directionrsquo

(lt Proto-Saami tā-mpē-l-tē [this-mpA-lA-ABL])

b) relational noun root + suffix -lA- + local case ending eg badje-l-is lsquofarther up higher

aboversquo (lt Proto-Saami pejē-lē-snē [above-lA-INE]) baji-l lsquofrom aboversquo (lt Proto-Saami

pejē-l-tē [above-lA-ABL])

Similar adverbs are found in many other Uralic languages as well As these kinds of forms

can be reconstructed on the basis of languages related to Finnic it is rather obvious that

similar formations must have also existed in Pre-Finnic at the time when uumll-postpositions

became grammaticalized as case endings During this suffixation process the endings of such

adverbs coincided with the newly emerged l-case endings In spite of this merger one can still

show that there are certain Finnic adverbs where a synchronic l-case ending probably does not

diachronically reflect an earlier uumll-postposition but a derivative in -lA instead

Finnic languages have adverbs in which a locative or separative case ending has been

added to a stem consisting of a pronoun root followed by a coaffix -kA- and the lsquolocalrsquo suffix

-l(A)- eg taumlaumlllauml lsquo(being) herersquo taumlaumlltauml lsquofrom herersquo (lt tauml-kauml-l-nauml tauml-kauml-l-tauml) siellauml

lsquo(being) therersquo sieltauml lsquofrom therersquo (lt si-kauml-l-nauml si-kauml-l-tauml) muualla lsquosomewhere else in

another placersquo muualta lsquofrom somewhere else from another placersquo (lt mū-ka-l-na mū-ka-

l-ta) The same suffixal combination -kA-lA- is found in derivatives with the suffix -inen

(eg taumlkaumllaumlinen lsquoa person from herersquo sikaumllaumlinen lsquoa person from therersquo muukalainen

lsquostrangerrsquo) and in such adverbs as mikaumlli lsquoif in the case thatrsquo and sikaumlli lsquoas far as in that

respectrsquo which have originally had a prolative meaning mikaumlli lsquothrough whatrsquo and sikaumlli

lsquothrough it that wayrsquo (Virtaranta 1962) Also series of postpositions and adverbs that have

been formed from relational noun roots with l-case endings are common eg sisaumlllauml lsquo(being)

inrsquo sisaumlltauml lsquo(coming) out from (the inside)rsquo sisaumllle lsquo(going) inrsquo edellauml lsquo(being) aheadrsquo edeltauml

lsquo(coming) from aheadrsquo edelle lsquo(going) aheadrsquo laumlhellauml lsquo(being) nearrsquo laumlheltauml lsquo(coming) from

near(by)rsquo laumlhelle lsquo(going) near (to)rsquo

The existence of these kinds of derivatives has frequently been thought to support the

lA-theory (see Section 2) It is indeed quite probable that many of them contain the Uralic

local derivational suffix -lA and in some cases the derived stem even has potential cognates

outside Finnic behind the series laumlhellauml laumlheltauml laumlhelle one can postulate the derived stem

laumlhe-l(auml)- lsquoplace nearbyrsquo which may be historically identical to Mari lišə-l lsquonear (ADJ)rsquo


132

Alhoniemi for instance has brought up the Mari derivatives with the suffix -l as an argument

supporting the lA-theory

Die [Tscheremissische] Staumlmme an die das Ableitungssuffix -l(V) tritt druumlcken durchweg ein spatiales

Verhaumlltnis aus lsquounter auf nahe fern neben uswrsquo Das an diese Worte tretende Suffix -l(V) brauchte also

nicht mehr die Lokalitaumlt auszudruumlcken sondern es konnte lsquodie Zugehoumlrigkeit zu der durch das Stammwort

ausgedruumlckten Lokalitaumlt od etwas daran Anschlieszligendesrsquo ausdruumlcken Da jedoch dieses spatiale Verhaumlltnis in

diesen Ausdruumlcken speziell eine aumluszligere Lokalitaumlt ist blieb die Bedeutung des Aumluszligeren natuumlrlich bei den lV-

Ableitungen und deren Flexionsformen erhalten Im Bewuszligtsein verknuumlpfte sich diese Bedeutung auch mit

dem Ableitungsuffix -lV Als sich aus diesen Ausdruumlcken des Aumluszligeren dann die zusammengesetzten

Kasusendungen mit -l zu entwickeln begannen war es natuumlrlich daszlig sie speziell die Bedeutung der aumluszligeren

Lokalitaumlt oder Habitivitaumlt erhielten wie es einerseits im Ostseefinnischen andererseits im Tscheremissischen

und Permischen geschehen ist (Alhoniemi 2001 109)

lsquoThe [Mari] stems which the derivational suffix -l(V) attaches to express essentially a spatial relationship

lsquounder on near far beside etcrsquo The suffix -l(V) that appears in these words did not need to express

locality any longer but it could express lsquothe affiliation to the locality expressed by the root word or

something connected to itrsquo Since however the spatial relationship in these expressions is particularly an

external location the meaning of the exterior was naturally obtained by the derivatives in lV and their

inflectional forms In the [speakersrsquo] consciousness this meaning became also attached to the derivational

suffix -lV As the compounded case endings with -l began to develop from these expressions of the exterior

it is natural that they retained particularly the sense of exterior locality and possession as it happened in

Finnic on the one hand and in Mari and Permic on the otherrsquo

This argumentation is quite impressionistic however even though semantic similarities are

pointed out there is no real attempt to explain how a morpheme that supposedly signified

something as vague as ldquothe affiliation to the locality expressed by the root word or something

connected to itrdquo would have developed the actually attested local functions of the Finnic l-

cases It is far from obvious what such an explanation could be as the concrete sense of

lsquolocation on the upper surfacersquo can be established as the core and primary function of these

cases within Finnic (see 31) It is true that derivatives consisting of a relational noun root and

a suffix -l(A) can to some extent be reconstructed on the basis of correspondences such as

Finnic laumlhe-l- ~ Mari lišə-l but nevertheless such derivatives do not offer any clear

evidence of the origin of the l-cases The semantics of such formations of relational noun

roots often do not agree with the specific sense of lsquolocation on the upper surfacersquo or even the

more general sense of lsquoexternal localityrsquo This is particularly obvious in the case of Finnish


133

sisaumlllauml ~ Livonian sizāl lsquo(being) inrsquo Finnish sisaumlltauml lsquo(coming) out from (the inside)rsquo Finnish

sisaumllle ~ Livonian sillotilde lsquo(going) inrsquo

One can add that if such archaic sets of derivatives really represented the diachronic

source of the Finnic l-cases it would be quite odd that precisely in these sets one encounters a

great deal of morphological variation and inconsistency For instance in modern Finnish the

l-case form sisaumllle lsquo(going) inrsquo is more or less in free variation with sisaumlaumln lsquo(going) inrsquo which

shows s-case morphology instead The situation with adverbs based on pronoun roots is even

more irregular the directional forms in these series usually show an entirely different suffix

-nnek eg taumlnne lsquo(coming) herersquo sinne lsquo(going) therersquo instead of expected taumlaumllle sielle

(lt tauml-kauml-llen si-kauml-llen) The directional form of muu- lsquootherrsquo has l-case morphology in

modern standard Finnish (muualle lsquo(going) somewhere elsersquo) but the form muuanne (lt mū-

ka-nnek) is attested in dialects The origin of the ending -nnek is unclear but in any case the

morphological irregularity of series of the type taumlauml-llauml taumlauml-ltauml tauml-nne is inconsistent with the

idea that these series represent the source of l-case endings It should be noted however that

this by no means excludes the possibility that such series may have exerted some secondary

influence on the development of these case forms as both the mentioned adverbs and the

nominal l-cases have coexisted in Finnic from its earliest stages on they have probably

affected each other ever since in ways that call for further research

In general one can say that the connection between the derivational suffix -lA and

adverbs of the type taumlaumlllauml lsquoherersquo and laumlhellauml lsquonearrsquo offers no counterargument for the uumll-

theory Assuming that l-cases developed through agglutination of original postpositions it is

only predictable that the newly emerged case endings coincided with various adverb endings

which originally contained the suffix -lA In this connection especially the postpositional

series paumlaumlllauml lsquoonLOCrsquo paumlaumlltauml lsquoonABLrsquo and paumlaumllle lsquoonLATrsquo is worth noting (cf 321) These

postpositions have traditionally been interpreted simply as l-case forms of the noun paumlauml

lsquohead endrsquo (SSA sv paumlauml Haumlkkinen 2004 sv paumlaumlllikkouml) which in turn goes back to Proto-

Uralic paumlŋi (Sammallahti 1988 548)

One can hypothesize however that also the paumlaumlllauml series could reflect an l-derivative

because the same element -l(l)- is also found in the prolative form paumlaumlllitse lsquooverrsquo and in

derivatives such as paumlaumll(l)ys lsquocoating cover(ing)rsquo paumlaumll(l)inen id paumlaumll(l)immaumlinen lsquotopmost

uppermostrsquo and paumlaumll(l)ikkouml lsquohead chiefrsquo Many such derivatives have a wide distribution in

Finnic languages and can be reconstructed for Proto-Finnic already cf Veps paumlľiči lsquooverrsquo

paumlluz lsquocoating cover(ing)rsquo paumlline id paumllembaine lsquotopmost uppermostrsquo and Estonian pealis

lsquocoating cover(ing)rsquo paumlaumlline lsquoid creamrsquo pealmine lsquotopmost uppermostrsquo (SSA sv paumlaumlllauml)


134

If paumlaumlllauml etc really originally were l-case forms of paumlauml lsquohead endrsquo one would have to think

that the -l- in the case endings would have become analogically reinterpreted as a part of the

stem in Proto-Finnic already and then spread to derivatives No clear parallels for such an

analogical change seem to be found however Instead it is worth noting that derivatives in -

lA such as sikaumllauml- lsquoplace therersquo and taumlkaumllauml- lsquoplace herersquo have corresponding prolative forms

with -l- especially in Karelian Lude and Veps eg Karelian mikaumllittši lsquoby whatrsquo sikaumllittši

lsquoby itrsquo taumlkaumllittši lsquoby thisrsquo (Suoniemi-Taipale 1994 134ndash135 154ndash155 161) Cognate forms

are attested even in Finnish albeit marginally (see also Virtaranta 1962 647ndash649)

(69) Oli=pa vauhtia Taumlmauml minun kansakoulukaverini

bePST3SG=DPT speedPTV this 1SGGEN elementaryschoolmate1SG

oli sikaumllitse eri maata ettauml haumln puki

bePST3SG insofar different countryPTV COMP 3SG dressPST3SG

samalla paumlaumllleen toiset housut - Tosin ei yhtauml

atthesametime paumlaumllle3SG anotherPL trousersPL tobesure NEG3SG as

nopeasti )

fastADV

lsquoWell that was fast This classmate of mine from the elementary school was different in

the sense that he put on another pair of trousers at the same time ndash Not that fast

though )rsquo (httpkeskustelusuomi24fi 1472006)

It is also interesting to note that the postpositions and adverbs in paumlaumll(l)- rather closely

resemble Saami adverbs and postpositions built from the Proto-Saami root pejē- with the

coaffix -l- cf eg North Saami bajil lsquofrom aboversquo badjel lsquooverrsquo badjelis lsquohigher up

higher above (LOC)rsquo badjelii lsquohigher up higher above (ILL)rsquo These Finnish and Saami word

families were etymologically equated by Rask (1832 37ndash38 see also Section 2) but in

modern etymological references the comparison is rejected due to irregular sound

correspondences The Proto-Saami form can be reconstructed as pejē-l(ē)- which would

presuppose a Finnic cognate of the shape pi(j)aumll- or puuml(j)aumll- not paumlaumll- In spite of this

irregularity the similarity is rather striking and it is tempting to assume that there could be a

historical connection between the two forms after all The idea receives some support from

the fact that there are derivatives which are widespread in both Finnic and Saami and which

share identical or similar morphology as shown in Table 12


135

Finnic languages Saami languages

Finnish paumlaumlllys Veps paumlluz Estonian

pealis lsquocoating cover(ing)rsquo

North Saami bajildus South Saami

bijjeldasse Skolt Saami pacircacirciacuteldotildes

lsquocoating cover(ing)rsquo

Finnic paumlaumlllekkaumlin Veps paumlleti lsquoon top of

each otherrsquo

North Saami badjaacutelaga(id) Skolt Saami

pacircjjlotildeotildeǥǥi ~ pacircjjlotildeotildežži lsquoon top of each

otherrsquo

Finnish paumlaumlllitse Vespian paumlliči lsquooverrsquo27 North Saami badjel lsquooverrsquo South Saami

bijjelen lsquoover ontorsquo Skolt Saami pacircacutejjel

lsquooverrsquo

Table 14 Some derivatives based on Finnic paumlaumll- and Saami pejē-l(ē)-

The cognation of the Finnic paumlaumll- and Saami pejē-l(ē)- is opposed by the irregular vowel

correspondence but the comparison could nevertheless be correct if the shape of the expected

Finnic reflex pi(j)aumll- would have been secondarily transformed to paumlaumll- due to

contamination with the noun paumlauml lsquohead endrsquo One could also think of another motive for the

irregular change the expected form pi(j)aumll- would have become very close or even identical

to another Finnish relational noun pieli lsquoedge sidersquo which according to Janhunen (1981

241) and Sammallahti (1988 539) goes back to Proto-Uralic pexli It is perhaps not

altogether irrelevant that in certain Finnic languages or dialects the reflexes of the expected

form pi(j)auml- would have completely merged with those of the noun paumlauml lsquohead endrsquo This is

the case in certain eastern dialects of Finnish and in Karelian where a diphthongization aumlauml gt

iauml took place cf piauml lsquohead endrsquo piaumlllauml lsquoon on top ofrsquo A similar diphthongization aumlauml gt ea

has also taken place in Estonian cf pea lsquohead endrsquo peal lsquoon on top ofrsquo The Estonian form

peal could also theoretically be a reflex of earlier pi(j)aumlllauml cf Estonian seal lsquotherersquo lt sial lt

sikaumlllauml (~ Finnish siellauml lsquotherersquo)

Due to phonological irregularities the equation of Finnic paumlaumll- and Saami pejē-l(ē)-

remains uncertain but the possibility should not be entirely rejected as is done by modern

etymological dictionaries (UEW 365 SSA sv paumlauml paumlaumlllauml Haumlkkinen 2004 sv paumlauml) ndash

especially when one takes into account that many other irregular and even downright 27 In this case the prolative suffix -itse -itši may be secondary as there are several cases where this suffix more

or less freely alternates with a shorter suffix -i cf Finnish ali ~ alitse lsquounderPROLrsquo yli ~ ylitse lsquooverPROLrsquo

laumlpi ~ laumlvitse lsquothroughrsquo


136

implausible etymological comparisons are accepted by the same dictionaries28 But whatever

the case the issue has no bearing on the origin of the Finnic l-cases regardless of whether the

Finnic paumlaumll- word family is originally based on l-case forms of paumlauml lsquohead endrsquo or an earlier

derivative paumlŋi-l(auml)- or pijauml-l(auml)- the l-case endings must still derive from Uralic uumll-

postpositions


As has been shown above the evidence presented for the earlier lA-theory is quite

unsystematic and insufficient Instead the correspondence between Finnic l-cases and Uralic

uumll-postpositions is rather striking and there is overwhelming evidence supporting their

cognation At this point then it is interesting to ponder why the lA-theory nevertheless

remained so widely accepted for almost a hundred years

Since the very beginning the lA-theory suffered from the basic weakness that it was

based on a kind of internal reconstruction of Finnish rather than a systematic comparison

between cognate languages After all the whole idea was originally based on Budenzrsquos

observation that the coaffix -l- resembled the suffix -lA and that in the phrase olla miehellauml

lsquoto be married (of a woman)rsquo the l-case form had a function comparable to a derivative in -lA

cf olla miehelaumlssauml id Even though additional arguments were presented by later scholars the

theory was never tested through systematic application of the comparative method One can

in fact say that Rask (1832 35ndash38) Donner (1879 84ndash93) and Bartens (2000 83) are the

only scholars who have chosen comparisons to other non-Finnic Uralic languages as a

starting point in their attempts to explain the development of the Finnic l-cases It is

furthermore interesting that only Rask managed to come close to the right solution even

though he naturally did not even have the chance to apply the comparative method which was

only developed decades later

Regardless of its weakness the lA-theory became widely accepted already in the early

20th century and it seems to have rather quickly turned into a piece of traditional academic

knowledge copied from one reference to the other This process can be understood in a wider

perspective on research history the tradition of research into Uralic historical morphology

must be criticized of the fact that the semantics and functions of morphemes have often not

28 For example UEW (365) compares Finnish paumlauml and its Uralic cognates (lt Proto-Uralic paumlŋi) to Old Turkish

maumlŋi lsquobrainrsquo and Mongolian heki lsquohead beginningrsquo


137

played a central role There has been a tendency to present hypotheses based primarily on the

phonological shape of the suffixes in question these kinds of comparisons are then supported

with references to vague similarities of meaning Thus also the ldquolocalrdquo l-cases in Finnic

languages came to be equated with the ldquolocalrdquo derivational suffix -lA

As another example of this research tradition one can mention the line of thought which

can be called lsquothe lative paradigmrsquo (see Footnote 20 in Section 34) In comparative Uralic

linguistics there is a tradition of reconstructing a variety of different directional case endings

so-called lsquolativesrsquo frequently suggested lative suffixes include at least -n -ń -ŋ -k -j

and -s Then numerous inflectional and derivational suffixes in the Uralic languages are

explained on the basis of these reconstructed latives One can distinguish between at least two

types of such lsquolative explanationsrsquo

a) The suffix is explained as a combination of two different lative suffixes mdash For example

the Finnic-Saami illative suffix -sin has often been thought to consist of a combination of

the lative suffixes -s and -n (eg Korhonen 1981 219) and the translative suffix -ksi

has in turn been explained on the basis of the lative suffixes -k and -s (eg Bartens

1999 77ndash78) In both cases i would be an epenthetic vowel which was added to avoid a

phonotactically illegal word-final consonant cluster

b) The suffix is explained as a combination of a lative (or several latives) and some other

suffix mdash It has been widely maintained that the endings of the inessive (-s-nA) and the

elative (-s-tA) are based to the lative -s after which the locative (-nA) and ablative (-

tA) suffixes were added (eg Korhonen 1981 222ndash224 see Ylikoski 2011 2016 for a

detailed critique of this tradition) As another example one may mention the Proto-Saami

modal suffix -ktē as in North Saami čehpe-t lsquoskillfullyrsquo (lt čeappē-ktē) larr čeahppi

lsquoskillfulrsquo it has been proposed that this consists of the lative -k and the ablative -tA

(Korhonen 1981 232ndash233) The Proto-Saami abessive suffix -ptākekn (which in North

Saami was degrammaticalized into the independent postposition haga) on the other hand

is considered to derive from the Proto-Uralic abessive suffix -ptA with two lative suffixes

added to it (ibid 226ndash227)

The lative paradigm however suffers from a fundamental weakness the presented

comparisons are nearly always semantically shallow and arbitrary There have been very few

serious attempts to show any functional connections between the various suffixes and the


138

purported ldquolativesrdquo underlying them to say nothing of actual attempts to reconstruct the

process of how the assumed combinations of ldquolativesrdquo and other suffixes arose and then

developed This is all the more problematic because the postulated combinations of suffixes

often appear strange when viewed from a functional perspective For example it is not easy to

see why the translative ending -ksi would be based on two conjoined directional case

endings29 One could hardly imagine for instance that the Finnish allative and illative case

endings could become conjoined in this manner and that the resulting combination would

then have a translative-like function as follows

(70) Haumln opiskeli opettaja-lle-seen

3SG study3SGPST teacher-ALL-ILL

lsquoShe studied to become a teacherrsquo

(pro opettaja-ksi [teacher-TRANSL])

29 It should be noted that more reasonable explanations of the origin of the translative ending -ksi have been

proposed as well Hakulinen (1979 101ndash102) considers the lative explanation methodologically dubious and

refers to Uotilarsquos (1945 335ff) view that the ending could be equated with homonymous derivational suffix -

ksi lsquomaterial for Xrsquo cf eg Finnish aida-kse-t lsquostakes (for building a fence)rsquo larr aita lsquofencersquo A particularly

plausible explanation is provided by Janhunenrsquos (1989 301) suggestion according to which the translative

ending derives from Proto-Uralic and is cognate with the Proto-Samoyed marker of the predestinative declension

-tə- This is reflected eg in Tundra Nenets -də- cf xərdeg lsquoknifersquo rarr xərdeg-də-da [knife-PREDES-3SG] xərdeg-də-

m-ta [knife-PREDES-ACC-3SG] xərdeg-də-n-ta [knife-PREDES-GEN-3SG] lsquoa knife for himrsquo (Salminen 1998 539)

Predestinative genitive forms come also functionally close to Finnish translatives as discussed by Salminen

(2014 289ndash294) and seen in (ivandashb)

(iv) a ťukudeg waeligsakoh ńe ńūm ńe-ddeg-n-ta medegda

this oldmanGEN woman child woman PREDES-GEN-3SG take3SGgtSG

b lsquoHaumln otti taumlmaumln ukon tyttaumlren vaimo-kse-nsarsquo

3SG takePST3SG thisGEN oldmanGEN daughterGEN wife-TRANSL-3SG

lsquoHe took that old manrsquos daughter as a wife for himrsquo (Tereščenko 1965 291 we are obliged to

Tapani Salminen for this example)

The sound correspondence between the suffixes -ksi and -tə- is entirely regular in Proto-Samoyed there was a

change -ks- gt -t- (cf eg Proto-Uralic miksa lsquoliverrsquo gt Proto-Samoyed mitə Janhunen 1981 251)


139

One has to stretch onersquos imagination even more to think of a combination of a directional and

a separative case ending in a modal function or that as many as two directional case endings

would be added after an abessive suffix

(71) Haumln opetti taitava-lle-lta ja jopa palka-tta-lle-lleseen

3SG teach3SGPST skillful-ALL-ABL and even salary-ABE-ALL-ALLILL

lsquoShe taught skillfully and even without salaryrsquo

(pro taitava-sti ja jopa palka-tta [skillful-ADV and even salary-ABE])

As the two pseudo-Finnish examples show the creation of new derivational suffixes or case

suffixes can hardly take place through mere unmotivated conjunction of two (or more)

existing case endings Such a development would be quite an extraordinary morphosyntactic

innovation and thus postulating that such an innovation has taken place in a reconstructed

proto-language ought to require quite extraordinary evidence as well (For more detailed

discussions on the few somewhat plausible instances of different kinds of case stacking in

Uralic see Ylikoski 2011 245ndash246 263 272 2016 36ndash41)

Whatever the actual background of the suffixes discussed above may be the loose

suppositions that connect them with various ldquolativesrdquo serve as a good examples of the flaws

of the lative paradigm the explanations offered for the origin of suffixes are generally

characterized by semantic opacity and absence of typological considerations (see also

Ylikoski 2016) What is more the phonological aspects of this method of explanation are also

unconvincing One merely needs to mechanically segment the suffixes and see if their

components could correspond to some other suffixes preferably to ldquolativesrdquo Vowels can

often be ignored as they can be explained away as epenthetic

illative -s-i-n = lative -s + epenthetic vowel + lative -n

translative -k-s-i = lative -k + lative -s + epenthetic vowel

modal suffix -k-tA = lative -k + ablative -tA

Regarding phonology it is crucial that only 17 consonant phonemes are reconstructed to

Proto-Uralic (Janhunen 1981 251 Sammallahti 1988 482) and five of these (c d ď r

x) seem to have been confined to lexical roots and are not known to have occurred in

suffixes Hence the six reconstructed lative suffixes -n -ń -ŋ -k -j and -s already

cover half of the consonants that can be found in any suffix When such an abundance of


140

phonological possibilities is combined with a nearly total lack of semantic constraints on the

comparisons it becomes an easy task indeed to discover ldquolativesrdquo wherever one looks for

The following thought experiment shows how seriously astray this kind of reasoning

may lead In the Eastern Finnmark dialects of North Saami the comitative plural ending -iguin

has become reduced to the form -jon ~ -jan as in ltmaacutenaacuteiguingt maumlaumlnaumlaumljon ~

maumlaumlnaumlaumljan lsquowith childrenrsquo In this case it is well-known that the suffix has developed

through the agglutination of an original postposition guoimme (see 35) but let us suppose

instead that we had instead merely reconstructed a comitative ending -jVn into a remote

proto-language In that case its postpositional background would obviously not be easy to

deduce but instead one could easily maintain that the suffix was a combination of the

ldquolativesrdquo -j and -n Given the vagueness of the semantic criteria generally applied in such

comparisons there should be no semantic objection to such an analysis after all also the

Finnish comitative-instructive (and genitive) ending -n has been considered related to the

ldquolativerdquo ending -n (eg Leino 2001)

We will mention yet one more particularly curious example of such weakly argued and

overoptimistic lative hypotheses There has even been an attempt to explain the development

of the Finnic l-cases on the basis of a lative Alvre (1986) argues that they are originally based

on Finno-Ugric lative suffix -l His main argument however is circular Alvre maintains

that because the s-cases have been explained on the basis of a lative suffix -s also the l-cases

can be best explained on the basis of a lative In addition to the general problems of the lative

paradigm such an induction is illogical Even if the s-cases were based on a lative suffix ndash

which has never been convincingly argued either (Ylikoski 2016) ndash this would still not reveal

anything about the origin of other case forms Alvre naturally tries to substantiate his

hypothesis by pointing out possible traces of this putative l-lative in various Finno-Ugric

languages but these comparisons are hardly convincing moreover he even resorts to

speculation with long-range comparisons to the Yukaghir instrumental suffix -le (which was

mentioned in 42) and certain suffixes in Tungusic languages

At this point it should be clear that one must categorically reject all morphological

explanations based on such random comparisons of suffixes with various kinds of ldquolativesrdquo

Instead Uralic historical morphology ought to start paying more attention to the functions of

morphemes It is not enough merely to explain the phonological shape of inflectional

morphemes in addition it is necessary to reconstruct the paths along which their usage has

developed This kind of explanation naturally requires painstaking application of the

comparative method to broad and representative sets of data and cannot be achieved by


141

superficial comparisons of morphemes and analysis of individual forms and construtions

selected at more or less random which has characterized much of the work done within the

context of the lative paradigm One can add that the typological knowledge we have today

offers a solid basis for the postulation of new hypotheses It is for instance well-known that

in the worldrsquos languages many suffixes have emerged through agglutination of originally

independent words but rarely indeed through the conjunction of various ldquolativesrdquo or other

directional case endings In Uralic linguistics these kinds of thoughts have been brought up

on a general level by Korhonen

As is known there are quite a number of rather young an[d] therefore transparent case forms derived from

postpositional constructions in the Uralic languages The postpositions from which the case suffixes originate

can mostly be traced back to nouns with concrete usually local or spatial meanings such as lsquothe insidersquo

lsquoupper sidersquo lsquobasersquo etc It also seems that case suffixes can originate from combinations of two or more

older case suffixes However some case suffixes that have traditionally been interpreted as suffix

combinations may with more thorough research prove to be original less transparent postpositions

(Korhonen 1991 177 emphasis added)

In this connection we can propose yet another new hypothesis inspired by this kind of

approach Earlier we mentioned the possibility that the possessive l-cases in Mari and Permic

languages might involve agglutinated cognates of postpositions based on a relational noun

root cognate with North Saami lu- Finnish luo- lsquoatrsquo (see 42) Ylikoski (2016) proposes that

the western Uralic s-cases as well the Samoyed local cases with the element -ntə-

may originate in Proto-Uralic postpositional phrases possibly based on a relational noun

seCV- (sekauml- seki- or sexi-) for lsquoinside interiorrsquo Another possible example of such

grammaticalization is the Proto-Finnic prolative ending -iccek (gt Finnish -itse) This does

not occur as a particularly productive case form in any Finnic language but it is found in

various adverbs such as Finnish maitse lsquoby landrsquo and meritse lsquoby searsquo No acceptable cognates

for this suffix have been shown from other Uralic languages (cf Suoniemi-Taipale 1994

230ndash247 Larjavaara 1995 613ndash615) Thus we propose that the prolative ending goes back to

a postposition śuumlδik which has a cognate in Saami North Saami čađa South Saami tjiumlrrh

Skolt Saami čotildeotildeđ lsquothroughrsquo etc (lt Proto-Saami čeδek) This word has been derived from

the same Uralic root as the noun śuumlδaumlmi lsquoheartrsquo (gt Skolt Saami čacircacircacuteđ Finnish sydaumln Mari

šuumlm Komi śelem Hungarian sziacutev etc lsquoheartrsquo) (SSA sv sydaumln UEW 477)

The equation of the prolative suffix -iccek with the postposition śuumlδik involves no

notable phonological problems The Pre-Proto-Finnic form of the suffix is reconstructed as -


142

ńćek or -ŋćek (cf Suoniemi-Taipale 1994 230ndash240 Larjavaara 1995 613ndash615) The Proto-

Finnic geminate affricate -cc- is apparently a product of secondary gemination a similar

development is also widely attested in the Proto-Finnic deminutive and adjective suffix -ise-

~ -icce- In modern Finnish the suffix mostly occurs in the form -(i)se- eg kala-nen

SGGEN kala-se-n lsquolittle fishrsquo villa-inen SGGEN villa-ise-n lsquowoollenrsquo In dialects one can find

vestigial forms pointing to a geminate affricate such as Tavastian Finnish semmo-tte-t lsquothose

kinds ofrsquo taumlmmouml-tte-t lsquothese kinds ofrsquo (-tt- lt -cc-) and in old literary Finnish such forms are

common eg Agricola synneitze-n lsquosinful-GENrsquo ~ modern Finnish syntise-n (Hakulinen 1979

124ndash125) The South Estonian cognate of this suffix also points to a geminate affricate as in

villa-notilde SGGEN villa-dsotilde SGILL villa-tsotilde-he lsquowoollenrsquo (~ Finnish villainen) (Keem 1997

32) The Pre-Proto-Finnic form of the suffix -ise- ~ -icce- can be reconstructed as -ńće-

which in turn derives from an even earlier form -nśi this is also the source of the Proto-

Saami deminutive suffix -ńče (gt North Saami -š -ž- eg kuolā-ńče gt guolaacute-š guolaacute-ž-

lsquolittle fishrsquo) (Sammallahti 1998 90)

The suffix -ise- ~ -icce- provides a good phonological parallel for the development of

the prolative suffix -iccek The oldest form of the suffix can be reconstructed as -nśik

which already comes close the postposition śuumlδik which can be reconstructed on the basis of

Saami The nasal -n- was originally the genitive ending on the complement of the

postposition and -śik can be quite naturally explained as a reduction of the form śuumlδik the

development would have been approximately meri-n śuumlδik gtgt merińśuumlik gtgt merińćik (gt

Finnish meritse lsquoby searsquo) Both the vowel uuml and the spirant δ are articulatorily weak sounds

and their loss in an unstressed position would be quite expected For example in the Eastern

Finnmark dialects of North Saami the phoneme δ shows the tendency to disappear between

unstressed vowels and hence forms such as ltboraditgt pooraδeh ~ pooraeh lsquoeat have a

mealrsquo are more or less in free variation The loss of the vowel uuml was already discussed in

Section 34 above

In addition to phonological arguments the equation of the Finnic prolative with the

(North) Saami postposition čađa naturally also requires the establishment of a semantic-

functional corresponce between these elements It is true the usage of the prolative does not

as exactly correspond to the postposition čađa as the usage of the l-cases does to the Saami

al-postpositions Instead the Finnic prolative ndash which is indeed not even a case form but

instead a weakly productive and rather rare type of adverb derivative ndash is often most naturally

translated into North Saami with the postposition bokte rather than čađa eg Finnish meritse

lsquoby searsquo = North Saami meara bokte One must note however that the meanings of the North


143

Saami postpositions bokte and čađa come rather close to each other Sammallahti (1998 232ndash

233) glosses them in English as lsquovia throughrsquo and lsquothroughrsquo respectively Moreover one can

indeed find a few prolative forms that can be translated exactly into North Saami with čađa

postpositional phrases and vice versa The following examples show that such

correspondences can be found in both traditional (72ndash73) and modern (74ndash75) functions of

the postposition čađa

(72) [ndash ndash] varsinkin jos kuluneilla sormilla on

especially if wearPSTPTCPPLADE fingerPLADE be3SG

vuosikausien turhana tyoumlnauml ollut killingin koumlyhaumln

multipleyearsPLGEN vainESS workESS bePSTPTCP coinGEN poorGEN

kuparin pyydystaumlminen joka saavuttamattomana on liukunut

copperGEN catchAN which unattainableESS be3SG slidePSTPTCP

koukistuvien raoitse niinkuin vesi seulan reijitse

bendPRSPTCPPLGEN gapPROL like water sieveGEN holePROL

lsquo[ndash ndash] eandalitge jos nohkan suorpmain leamaš jahkemeriid

especially if wearPSTPTCP fingerPLLOC bePSTPTCP multipleyears

duššibargun haacutehpohallat vaacuteivvaacuteš veaikešilliŋggaid mat

vainworkESS gropeINF poor coppershillingPLGA whichPL

juksameahttumin leat johtaacuten suorbmalanjaid čađa dego čaacutehci

unattainableESS be3PL slipPSTPTCP fingergapPLGA čađa like water

silleraacuteiggiid čađarsquo

sieveholePLGA čađa

lsquo[ndash ndash] especially if onersquos worn fingers have for years been grasping in vain for scanty

copper shillings that have unattainably slipped through the gaps of onersquos crooked

fingers like water through holes of a sieversquo (Kilpi 1993 [1933] 121)

(73) Aumlaumlni laumlheni laumlhenemistaumlaumln sillauml talvitie kulki

sound comenearPST3SG comenearANELA3SG because winterway goPST3SG

Telkiaumln pihatse

TelkiaumlGEN yardPROL

lsquoJietna lahkonii ahtrsquo lahkonii dasgo daacutelvemaacuteđii

sound comenearPST3SG COMP comenearPST3SG because winterway

manai Telkiauml šilju čađarsquo


144

goPST3SG TelkiaumlGA yardGA čađa

lsquoThe sound came nearer and nearer as the winter way went through Telkiaumlrsquos yardrsquo

(Reijonen 1900 427ndash428)

(74) [N N] lea vaacuteidaacuten Guovdageainnu lagasraacutedio (GLR)

N N be3SG accusePSTPTCP GuovdageaidnuGA localradioGA (GLR)

leansmaacutennii Daningo dikte muhtun boazosaacutepmelačča soaibmat

lensmannILL because allowPST3PL some SaamireindeerherderGA revileINF

su raacutedio čađa

3SGGA radioGA čađa

lsquoN N on tehnyt Koutokeinon paikallisradiosta (GLR)

N N be3SG makePSTPTCP GuovdageaidnuGEN localradioELA (GLR)

valituksen nimismiehelle koska eraumlaumln porosaamelaisen

complaintGEN lensmannALL because certain SaamireindeerherderGEN

sallittiin haukkua haumlntauml radioitsersquo

allowPSTPASS revileINF 3SGPTV radioPROL

lsquoN N has filed a complaint against the Guovdageaidnu local radio station (GLR) at the

Lensmann because one Saami reindeer herder was allowed to revile him on the radiorsquo

(MAacute 1995)

(75) Jearahallan 51 čearuid saacutegadoalliid gaskkas

surveyAN 51(GA) reindeerherdingdistrictPLGA chairpersonPLGA among

vuoseha ahte stuorimus vaacutettisvuohta lea oalle heajos vejolašvuođat

show3SG COMP greatSUP difficulty be3SG rather bad possibilityPL

gulahallat telefuvnnaid čađa

communicateINF telephonePLGA čađa

lsquoKysely 51 paliskunnan puheenjohtajien keskuudessa

survey 51(GEN) reindeerherdingdistrictGEN chairpersonPLGA among

osoittaa ettauml suurin ongelma ovat sangen huonot mahdollisuudet

show3SG COMP greatSUP problem be3PL rather badPL possibilityPL

keskustella puhelimitsersquo

communicateINF telephonePROL


145

lsquoA survey of chairpersons of 51 reindeer herding districts shows that the greatest

difficulty is posed by the very poor possibilities to communicate by telephonersquo

(httpwwwglesbygdsverketse 1042007)

The uumll-theory presented in this paper and the new hypothesis of the origin of the Finnic

prolative serve as examples of what kind of insights more remotely related Uralic languages

can offer to the study of Finnic historical morphology It was after all a fatal weakness of the

earlier lA-theory that it did not take evidence from languages outside the Finnic group into

serious consideration Because of this it is regrettable that it has already become a sort of a

tradition to examine the history of Finnish from a narrow language-internal perspective

diachronic hypotheses are often based on material collected from Finnish exclusively often

even neglecting material from other closely related Finnic languages (eg Inaba 2002 254ndash

261 and Ylikoski 2005 have noted that this kind of argumentation is becoming widespread)

But when the study of the history of Finnish and Finnic is correctly viewed as one subfield of

Uralic historical linguistics the background of many linguistic phenomena in Finnish reveal

themselves in an altogether different light

Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal

APPR1 first approximative (case)

ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive

DPT discourse particle

DU dual

ELA elative

ESS essive

EX existential

FUT future


146

GA genitive-accusative

GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural

PREDES predestinative

PROG progressive

PROL prolative

PRS present tense

PST past tense

PST2 second past tense

PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material

Aikio Annukka amp Aikio Samuli 1978a Girdinoaiddi baacuterdni Saacutepmelaš maacuteidnasat Porvoo ndash

Helsinki ndash Juva WSOY

ndashndashndash 1978b Lentonoidan poika Saamelaisia satuja Porvoo ndash Helsinki ndash Juva WSOY

Blind Ella Karin 1992 Eallima govat Johkamohkki Saacutemi Girjjit

Castreacuten Matthias Alexander 2005 Saacutemi maacutetkkit 1838 ja 1841ndash42 Kaacuteraacutešjohka

ČaacutelliidLaacutegaacutedus

Gaski Harald amp Solbakk John T amp Solbakk Aage (eds) 2004 Min njaacutelmmaacutelaš aacuterbevierru

Maacuteidnasat myhtat ja muitalusat [Kaacuteraacutešjohka] Davvi Girji os

Guttorm Eino 1981 Aacuterbeeatnan luohti Deatnu Jaringrrsquogalaeligdrsquodji Arings

Haeligtta Lars ndash Baeligr Anders 1982 [1958] Muitalusat Tromsoslash ndash Oslo ndash Bergen

Universitetsforlaget


147

ndashndashndash 1993 Usko ja elaumlmauml Koutokeinon saamelaisten hengellisestauml elaumlmaumlstauml Lars Levi

Laestadiuksen heraumlyksestauml ja lestadiolaisuuden alkuvaiheista ennen vuotta 1852

Utsjoki Girjegiisaacute

Jansson Tove 1979 [1965] Muumipappa ja meri Porvoo ndash Helsinki ndash Juva WSOY

Jansson Tove 1990 Aacutehčči ja mearra Ohcejohka Girjegiisaacute

Kristiansen Roald E 2004a Davviguovlluid eamiaacutelbmogiid oskkut ja saacutemi dološ osku

Oahppogirji nuoraidskuvlla 8ndash10 luohkaacuteide Kaacuteraacutešjohka ČaacutelliidLaacutegaacutedus

ndashndashndash 2004b Leastadianisma Oahppogirji nuoraidskuvlii Kaacuteraacutešjohka ČaacutelliidLaacutegaacutedus

Marastat Mihkkal 1990 Maacutehkanvaacuteri gumppet Guovdageaidnu DAT

ndashndashndash 1991 Siidavuomi golli Guovdageaidnu DAT

ndashndashndash 1992 Darjeskaacuteiddi čiehkaacute Guovdageaidnu DAT

MAacute = Min Aacuteigi [Approximately 150 issues published in years 1995 and 1997 provided to the

Giellagas Institute for Saami Studies (University of Oulu Finland) by the Language

Bank project of the Nordic Sami Institute (Guovdageaidnu Norway)] Kaacuteraacutešjohka

Mukka Timo K 1966 Laulu Sipirjan lapsista Jyvaumlskylauml Gummerus

ndashndashndash 2005 Sipirjaacute Kaacuteraacutesjohka Davvi Girji os

OT = Ođđa Testamenta Oslo Norgga Biibbalsearvi 1998

Piibel = Piibel Vana ja Uus Testament Tallinn Eesti Piibliselts 1997

Raamattu = Raamattu Suomen evankelis-luterilaisen kirkon kirkolliskokouksen vuonna 1992

kaumlyttoumloumln ottama suomennos lthttpwwwevlfiraamattu1992gt

Sombi Aacutesllat 1996 Jaacutevvaacutesan goahtesajit Daacutehpaacutehusat dološ badjeeallimis Kaacuteraacutešjohka

Davvi Girji os

Turi Klemet Nilsen 1982 Aacuteiggit rivdet [Deatnu] Jaringrrsquogalaeligdrsquodji Arings

US = Uuzi Sana Koitepainos Helsinki Biblienkiaumlnnaumlnduumlinstituuttu 2003

UT = Ūž Testament Helsinki Suomalaisen Kirjallisuuden Seuran Kirjapainon Oy 1942

Vars Ellen Marie 1990 Arvedaacutevggi maacutenaacutet [Kaacuteraacutešjohka] Davvi Girji os

Vest Jovnna-Aacutende 1988 Čaacutehcegaacuteddaacutei nohkaacute boazobaacutelggis Kaacuteraacutešjohka Davvi Girji os

ndashndashndash 1990 Poropolku sammaloituu Oulu Pohjoinen

VS = Выль Сӥзён Стокгольм ndash Хельсинки Библиез берыктонъя Институт 1997

AringT = Aringdaring Testamennta Uppsala Svieriga raacutemaacutetsiebrre 2000


148

Other sources of material

Biblia 1642 = Vanhan testamentin II osa Joosuan kirja Tuomarien kirja Ruutin kirja

Samuelin kirjat Kuningasten kirjat Aikakirjat Esran kirja Nehemian kirja ja Esterin

kirja [text corpus] ndash Helsinki Institute for the Languages of Finland

lthttpkainokotusfikorpusvksmetabibliavt2_joos_rdfxmlgt 1342007

Cuningsen Maijtin Saumlaumlndouml ja Asetus Palcollisist ja Palckawaumlest 1723 = 1700-luvun

asetustekstejauml [text corpus] ndash Helsinki Institute for the Languages of Finland

lthttpkainokotusfikorpusvksmetalaitas1700_rdfxmlgt 242007

IK = Itkonen Erkki 1992 Inarinsaamelaisia kielennaumlytteitauml Aanaarkiela čȧjttuzeh Ed by

Lea Laitinen Memoires de la Socieacuteteacute Finno-Ougrienne 213 Helsinki Socieacuteteacute Finno-

Ougrienne

Kilpi Volter 1993 [1933] Alastalon salissa Kuvaus saaristosta II Helsinki Otava

LAFD = Lexical Archive of Finnish Dialects (Suomen murteiden sana-arkisto) Institute for

the Languages of Finland (Kotus) Helsinki

LS = Bull Ella Holm amp Bergsland Knut 1993 Lohkede Saemien Soslashrsamisk lesebok

Kaacuteraacutešjohka Davvi Girji os

OTGr = The NRSVndashNIVtrade parallel New Testament in Greek and English With interlinear

translation by Alfred Marshall Grand Rapids Michigan Zondervan Publishing House

1990

Reijonen Juho 1900 Kertoelmia ja kuvauksia Porvoo Werner Soumlderstroumlm

Sammallahti Pekka (ed) 2004 Javrij jieŋah parguu Anaacuterašgiel lohkosat Publications of the

Giellagas Institute 2 Oulu University of Oulu

ndashndashndash (2012) Vuotildeacutelǧǧe jaringaringacutetted ooudacircs Saumlaumlacutemǩiotildellsaž lookkacircmǩeacuterjj Publications of the


Suomenkieliset Tieto-Sanomat 171776 = Lizelius Antti Suomalaiset Tieto-Sanomat 1775 ja

1776 [text corpus] ndash Helsinki Institute for the Languages of Finland

lthttpkainokotusfikorpusvksmetalizeliussts1775_rdfxmlgt 242007

Vanhan kirjasuomen korpus Helsinki Institute for the Languages of Finland

lthttpkainokotusfikorpusvksmetavks_coll_rdfxmlgt 242007


149

Other references

Ahlqvist August 1863 Om Ungerska spraringkets foumlrvandtskap med Finskan ndash Suomi II 1 1ndash

60 Helsinki

ndashndashndash 1877 Suomen kielen rakennus Vertaavia kieliopillisia tutkimuksia I Nominien Synty ja

Taivutus Suomalainen Runo-oppi Helsinki Suomalaisen Kirjallisuuden Seura

Aikio Ante 2014 The Uralic-Yukaghir lexical correspondences genetic inheritance

language contact or chance resemblance ndash Finnisch-Ugrische Forschungen 62 7ndash76

Aikio Ante amp Ylikoski Jussi 2007 Suopmelaš gielaid l-kaacutesusiid aacutelgovuođđu saacuteme- ja earaacute

fuolkegielaid čuovggas ndash Jussi Ylikoski amp Ante Aikio (eds) Saacutemit saacutenit saacutetnehaacutemit

Riepmočaacutela Pekka Sammallahtii miessemaacutenu 21 beaivve 2007 Suomalais-Ugrilaisen

Seuran Toimituksia 253 Helsinki Suomalais-Ugrilainen Seura 11ndash71

Alhoniemi Alho 1979 Suomen kielen l- ja s-sijojen oppositiosta ndash Kaisa Haumlkkinen Jussi

Kallio amp Leena Kytoumlmaumlki (eds) Sanomia Juhlakirja Eeva Kangasmaa-Minnin 60-

vuotispaumlivaumlksi 1441979 Turun yliopiston suomalaisen ja yleisen kielitieteen laitoksen

julkaisuja 9 Turku Turun yliopisto 89ndash105

ndashndashndash 1985 Marin kielioppi Apuneuvoja suomalais-ugrilaisten kielten opintoja varten X

Helsinki Suomalais-Ugrilainen Seura

ndashndashndash 2001 Uumlber den alten und neuen Lokalkasussuffixe im Tscheremissischen ndash Finnisch-

Ugrische Forschungen 56 95ndash115

Alvre Paul 1986 Zu den finnisch-ugrischen l-Kasus ndash Советское финно-угроведение 22

81ndash87

Anttila Raimo amp Uotila Eeva 1984 Finnish ovela lsquosly cunningrsquo and the Baltic Finnic outer

local cases ndash Ural-Altaische Jahrbuumlcher 56 121ndash128

Baker Robin 1985 The development of the Komi case system A dialectological

investigation Suomalais-Ugrilaisen Seuran Toimituksia 189 Helsinki Suomalais-

Ugrilainen Seura

Bartens Raija 1978 Synteettiset ja analyyttiset rakenteet lapin paikanilmauksissa

Suomalais-Ugrilaisen Seuran Toimituksia 166 Helsinki Suomalais-Ugrilainen Seura

ndashndashndash 1999 Mordvalaiskielten rakenne ja kehitys Suomalais-Ugrilaisen Seuran Toimituksia

232 Helsinki Suomalais-Ugrilainen Seura

ndashndashndash 2000 Permilaumlisten kielten rakenne ja kehitys Suomalais-Ugrilaisen Seuran Toimituksia



150

Batalova = Баталова Р М 1982 Ареальные исследования по восточным финно-

угорским языкам (коми языки) Москва Наука

Bergsland Knut 1946 Roslashros-lappisk grammatikk Et forsoslashk paring strukturell spraringkbeskrivelse

Instituttet for sammenlignende kulturforskning Serie B Skrifter XLIII Oslo H

Aschehoug amp Co

Blake Barry J 2001 Case Cambridge Cambridge University Press

Blomstedt Oskar Aukusti Frithiof 1869 Halotti Beszeacuted ynnauml sen johdosta Wertailevia

Tutkimuksia Unkarin Suomen ja Lapin kielissauml Helsinki J Kr Frenckell

Budenz Joacutezsef 1886 Az ugor nyelvek oumlsszehasonliacutetoacute alaktana Első reacutesz Az ugor nyelvek

szoacutekeacutepzeacutese II Neacutevszoacutekeacutepzeacutes ndash Nyelvtudomaacutenyi Koumlzlemeacutenyek 20 401ndash474

Castr[eacute]n Matthias Al[e]xander 1839 De affinitate declinationum in lingua Fennica

Esthonica et Lapponica Helsingforsiae Typis Frenckellianis

Castreacuten Matthias Alexander 1844 Elementa grammatices Syrjaenae Helsingforsiae Ex

officina typographica heredum Simelii

ndashndashndash 1854 = M Alexander Castreacutenrsquos Grammatik der samojedischen Sprachen Im Auftrage

der Kaiserlichen Akademie der Wissenschaften herausgegeben von Anton Schiefner St

Petersburg Buchdruckerei der Kaiserlichen Akademie der Wissenschaften

ndashndashndash 1858 [1849] = M Alexander Castreacutenrsquos Versuch einer ostjakischen Sprachlehre nebst

kurzem Woumlrterverzeichniss Herausgegeben von Anton Schiefner Zweite verbesserte

Auflage St Petersburg Buchdruckerei der Kaiserlichen Akademie der Wissenschaften

Collinder Bjoumlrn 1952 Uralaltaisch ndash Ural-Altaische Jahrbuumlcher 24 3ndash4 1ndash26

ndashndashndash 1960 Comparative grammar of the Uralic languages Stockholm Almqvist amp Wiksell

Donner O 1879 Die gegenseitige Verwandtschaft der finnisch-ugrischen Sprachen

Abdruck aus den Acta Soc Scient Fennicaelig Tom XI Helsingfors Druckerei der

Finnischen Litteratur-Gesellschaft

Friis J A 1856 Lappisk Grammatik Udarbeidet efter den finmarkiske Hoveddialekt eller

Sproget saaledes som det almindeligst tales i norsk Finmarken Christiania J W

Cappelen

Genetz Arvid 1896 Ensi tavuun vokaalit suomen lapin ja mordvan kaksi- ja

useampitavuisissa sanoissa Helsinki Suomalaisen kirjallisuuden seura

Givoacuten Talmy 1979 Ute dictionary Preliminary edition Ignacio Colorado Ute Press

ndashndashndash 1980 Ute reference grammar First edition Ignacio Colorado Ute Press

Haumlkkinen Kaisa 1983 Suomen kielen vanhimmasta sanastosta ja sen tutkimisesta

Suomalais-ugrilaisten kielten etymologisen tutkimuksen perusteita ja metodiikkaa


151

Turun yliopiston suomalaisen ja yleisen kielitieteen laitoksen julkaisuja 17 Turku

Turun yliopisto

ndashndashndash 1984 Waumlre es schon an der Zeit den Stammbaum zu faumlllen ndash Ural-Altaische

Jahrbuumlcher Neue Folge 4 1ndash24

ndashndashndash 1985 Suomen kielen aumlaumlnne- ja muotorakenteen historiallista taustaa Fennistica 6

Turku Aringbo Akademi

ndashndashndash 2002 Suomen kielen historia 1 Suomen kielen aumlaumlnne- ja muotorakenteen historiallista

taustaa Turun yliopiston suomalaisen ja yleisen kielitieteen laitoksen julkaisuja 69

Turku Turun yliopisto

ndashndashndash 2004 Nykysuomen etymologinen sanakirja Juva WSOY

Hakulinen Lauri 1941 Suomen kielen rakenne ja kehitys Ensimmaumlinen osa Aumlaumlnne- ja

muoto-oppia Helsinki Otava

ndashndashndash 1979 Suomen kielen rakenne ja kehitys Neljaumls korjattu ja lisaumltty painos Helsinki

Otava

Halling Tiina 1996 Vaihtoehtoja ulkopaikallissijoille liivissauml ndash Heikki Leskinen Saacutendor

Maticsaacutek amp Totildenu Seilenthal (eds) Congressus Octavus Internationalis Fenno-

Ugristarum Jyvaumlskylauml 10ndash15 8 1995 Pars III Sessiones sectionum Phonologia amp

Morphologia Jyvaumlskylauml 100ndash104

ndashndashndash 1999 About Livonian preposition iļ ndash Ago Kuumlnnap (ed) Indo-European-Uralic-

Siberian linguistic and cultural contacts Fenno-Ugristica 22 Tartu 65ndash76

Honti Laacuteszloacute 2006 Eraumlaumlstauml ugrilaisten kielten postpositioperaumlisestauml kaasussuffiksien

perheestauml ndash Suomalais-Ugrilaisen Seuran Aikakauskirja 91 81ndash91

Hunfalvy Paacutel 1864 A vogul foumlld eacutes neacutep Reguly Antal hagyomaacutenyaiboacutel kidolgozta Hunfalvy

Paacutel Pest Eggenberger Ferdinaacutend Akademiai Koumlnyvaacuterus

Huumo Tuomas 1995 Paikallissijan kieliopillistuminen datiivi-genetiivin funktioon

Uralilaisen n-sijan ja itaumlmerensuomalaisen adessiivin kehityksen vertailua ndash Sananjalka

37 55ndash79

Huumo Tuomas amp Ojutkangas Krista 2006 An introduction to Finnish spatial relations

Local cases and adpositions ndash Marja-Liisa Helasvuo amp Lyle Campbell (eds) Grammar

from the human perspective Case space and person in Finnish 11ndash20

Inaba Nobufumi 2001 Elollistarkoitteinen laumlhde ja sen s-sijainen merkintauml

itaumlmerensuomalaisissa kielissauml sijanmerkinnaumln ja sijajaumlrjestelmaumln suhteesta II ndash

Sananjalka 43 26ndash64


152

ndashndashndash 2002 Ensimmaumlisen suomenkielisen Biblian allatiivin kaumlyttouml modernin kuvausvaumllineen

valossa ndash Sananjalka 44 247ndash262

Itkonen Erkki 1957a Lokatiivista ja genetiivistauml ym ndash Virittaumljauml 61 308ndash320

ndashndashndash 1957b Loppusanat prof Kettuselle ndash Virittaumljauml 61 435ndash439

ndashndashndash 1966 Kieli ja sen tutkimus Helsinki WSOY

IW = Itkonen Erkki 1986ndash1991 Inarilappisches Woumlrterbuch Lexica Societatis Fenno-

Ugricae XX Helsinki Suomalais-Ugrilainen Seura

Janhunen Juha 1977 Samojedischer Wortschatz Gemeinsamojedische Etymologien

Castrenianumin toimitteita 17 Helsinki

ndashndashndash 1981 Uralilaisen kantakielen sanastosta ndash Suomalais-Ugrilaisen Seuran Aikakauskirja

77 219ndash274

ndashndashndash 1982 On the structure of Proto-Uralic ndash Finnisch-Ugrische Forschungen 44 23ndash42

ndashndashndash 1989 Samojedin predestinatiivisen deklinaation alkuperaumlstauml ndash Suomalais-Ugrilaisen

Seuran Aikakauskirja 82 298ndash301

ndashndashndash 1998 Samoyedic ndash Daniel Abondolo (ed) The Uralic languages London ndash New York

Routledge 457ndash479

Kallio Petri 2012a Jaumllkitavujen diftongit kantasuomessa ndash Fenno-Ugrica Suecana Nova

Series 14 31ndash40

ndashndashndash 2012b The prehistoric Germanic loanword strata in Finnic ndash Riho Gruumlnthal amp Petri

Kallio (eds) A linguistic map of prehistoric Northern Europe Meacutemoires de la Socieacuteteacute

Finno-Ougrienne 266 Helsinki Socieacuteteacute Finno-Ougrienne 225ndash238

Keem Hella 1997 Votilderu keel Tallinn Eesti Teaduste Akadeemia Emakeele Selts ndash Votildero

Instituut

Kettunen Lauri 1943 Vepsaumln murteiden lauseopillinen tutkimus Suomalais-Ugrilaisen

Seuran Toimituksia 86 Helsinki Suomalais-Ugrilainen Seura

ndashndashndash 1957 Vastinetta Erkki Itkoselle ndash Virittaumljauml 61 428ndash434

KKS = Virtaranta Pertti amp Raija Koponen (eds) 1968ndash2005 Karjalan kielen sanakirja

Lexica Societatis Fenno-Ugricae XVI Kotimaisten kielten tutkimuskeskuksen

julkaisuja 25 Helsinki Suomalais-Ugrilainen Seura ndash Kotimaisten kielten

tutkimuskeskus

Korhonen Mikko 1979 Entwicklungstendenzen des finnisch-ugrischen Kasussystems ndash

Finnisch-Ugrische Forschungen 43 1ndash21

ndashndashndash 1981 Johdatus lapin kielen historiaan Suomalaisen Kirjallisuuden Seuran toimituksia

370 Helsinki Suomalaisen Kirjallisuuden Seura


153

ndashndashndash 1991 Remarks on the structure and history of the Uralic case system ndash Suomalais-

Ugrilaisen Seuran Aikakauskirja 83 163ndash180

Kracht Marcus 2005 The Semantics of Locatives in the Uralic Languages ndash Jocelyne

Fernandez-Vest (ed) Les Langues Ouraliennes aujourdhui Bibliothegraveque de lEcole

des Hautes-Etudes Sciences Historiques et Philologiques No 340 Editions Honoreacute

Champion 145ndash158

Kulonen Ulla-Maija 1993 Johdatus unkarin kielen historiaan Suomi 170 Helsinki

Suomalaisen Kirjallisuuden Seura

Laaksonen Heli 2000 Adessiivi suomessa ja virossa ndash funktiot frekvenssit yhtaumllaumlisyydet

erot vastineet Pro gradu -tutkielma Turun yliopiston suomen kielen ja yleisen

kielitieteen laitos Turku

Laanest Arvo 1982 Einfuumlhrung in die ostseefinnischen Sprachen Autorisierte Uumlbertragung

aus dem Estnischen von Hans-Hermann Bartens Hamburg Helmut Buske Verlag

Larjavaara Matti 1995 Prolatiivi ennen ja nyt ndash Virittaumljauml 99 609ndash616

Lauranto Yrjouml 1994 Sisauml- ja ulkopaikallissijat onko perinteinen opettamisjaumlrjestys

perusteltu ndash Minna Suni amp Eija Aalto (eds) Suuntaa suomenopetukseen ndash tuntumaa

tutkimukseen Korkeakoulujen kielikeskuksen selosteita 4 Jyvaumlskylauml Korkeakoulujen

kielikeskus 37ndash61

Laver John 1994 Principles of phonetics Cambridge Cambridge University Press

Leem Knud 1748 En lappisk Grammatica efter den Dialect som bruges af Field-Lapperne

udi Porsanger-Fiorden Kioslashbenhavn Gottman Friderich Kisel

Lehtisalo T 1936 Uumlber die primaumlren ururalischen Ableitungssuffixe Suomalais-Ugrilaisen


ndashndashndash 1956 Juraksamojedisches Woumlrterbuch Lexica Societatis Fenno-Ugricae XIII Helsinki

Suomalais-Ugrilainen Seura

Leino Pentti 1989 Paikallissijat ja suhdesaumlaumlntouml kognitiivisen kieliopin naumlkoumlkulma ndash

Virittaumljauml 93 161ndash219

ndashndashndash 1990 Spatial relations in Finnish a cognitive perspective ndash Ingrid Almqvist Per-Erik

Cederholm amp Jarmo Lainio (eds) Fraringn Pohjolas poumlrten till kognitiv kontakt Vaumlnskrift

till Erling Wande den 9 maj 1990 Stockholm studies in Finnish language and literature

6 Stockholm 117ndash152

ndashndashndash 2001 Henkiloumlviitteinen allatiivi ndash Leino Pentti amp Herlin Ilona amp Honkanen Suvi amp

Kotilainen Lari amp Leino Jaakko amp Vilkkumaa Maija 2001 Roolit ja rakenteet


154

Henkiloumlviitteinen allatiivi Biblian verbikonstruktioissa Suomalaisen Kirjallisuuden

Seuran toimituksia 813 Helsinki Suomalaisen Kirjallisuuden Seura

Levinson Stephen C 2003 Space in language and cognition Explorations in cognitive

diversity Cambridge Cambridge University Press

Loumlnnrot Elias 1841 Bidrag till Finska Spraringkets Grammatik (Forts fr 4de Haumlftet) ndash Suomi

I 5 29ndash47 Helsingfors

LW = Kettunen Lauri 1938 Livisches Woumlrterbuch mit grammatischer Einleitung Lexica

Societatis Fenno-Ugricae V Helsinki Suomalais-Ugrilainen Seura

Maumlgiste Julius 1928 Eraumlitauml liivilaumlisiauml possessiivisuffiksin jaumlaumlnnoumlksiauml ndash Virittaumljauml 32 285ndash

287

Maslova Elena 2003 A Grammar of Kolyma Yukaghir Mouton Grammar Library 27 Berlin

ndash New York Mouton de Gruyter

Merimaa Heidi 2002 Adpositiot suomen vanhoissa kieliopeissa Pro gradu -tutkielma Turun

yliopiston suomalaisen ja yleisen kielitieteen laitos Turku

Mikola Tibor 1975 Die alten Postpositionen des Nenzischen (Juraksamojedischen) Den

Haag ndash Paris Mouton Budapest Akadeacutemiai Kiadoacute

Moisio Arto amp Saarinen Sirkka 2008 Tscheremissisches Woumlrterbuch Aufgezeichnet von

Volmari Porkka Arvid Genetz Yrjouml Wichmann Martti Raumlsaumlnen T E Uotila und Erkki

Itkonen Lexica Societatis Fenno-Ugricae 32 Kotimaisten kielten tutkimuskeskuksen


tutkimuskeskus

MW = H Paasonens Mordwinisches Woumlrterbuch Zusammengestellt von Kaino Heikkilauml

Bearbeitet und herausgegeben von Martti Kahla Lexica Societatis Fenno-Ugricae

XXIII Helsinki Suomalais-Ugrilainen Seura 1990ndash1999

Nickel Klaus Peter 1994 Samisk grammatikk Kaacuteraacutešjohka Davvi Girji

Nielsen Konrad 1979 [1932ndash1962] Lappisk (samisk) ordbok Grunnet paring dialektene i

Polmak Karasjok og Kautokeino Oslo Universitetsforlaget

Niemi Jaana amp Mosin Mihail 1995 Ersaumllaumlis-suomalainen sanakirja Turun yliopiston

suomalaisen ja yleisen kielitieteen laitoksen julkaisuja 48 Turku Turun yliopisto

Nikolaeva Irina 2006 A Historical Dictionary of Yukaghir Berlin ndash New York Mouton de

Gruyter

Nyikolajeva Irina 2000 Chrestomathia jucagirica Uraacutelisztikai Tanulmaacutenyok 10 Budapest

ELTE BTK Finnugor Tanszeacutek


155

Oinas Felix J 1961 The development of some postpositional cases in Balto-Finnic

languages MSFOu 123

Ojutkangas Krista 2005 Viittauskehykset ja tarkastelunaumlkoumlkulma ndash miten sijaintia

perusakseleilla kuvataan ndash Virittaumljauml 109 525ndash551

Pajusalu Elna 1957a Soome-ugri l-kaumlaumlnetest ndash Emakeele Seltsi aastaraamat 3 159ndash172

ndashndashndash 1957b Uumllevaade laumlaumlnemere keelte ablatiivi funktsioonidest ndash Eesti NSV Teaduste

Akadeemia Toimetised VI Uumlhiskonnateaduste seeria 2 133ndash153

ndashndashndash 1958a Adessiivi funktsioonid eesti murretes ja laumlhemates sugulaskeeltes ndash Keel ja

Kirjandus 4ndash51958 246ndash258

Pajusalu = Паюсалу Э 1958 Внешнеместные падежи в прибалтийско-финских языках

(функции падежей) Автореферат диссертация на соискание ученой степени

кандидата филологических наук Таллин Академия наук Эстонской ССР

Институт языка и литературы

Pajusalu Elna 1960 Laumlaumlnemere keelte allatiivi funktsioonid ndash Eesti NSV Teaduste

Akadeemia Keele ja Kirjanduse Instituudi uurimused 5 88ndash116

Papp Istvaacuten 1968 Unkarin kielen historia Tietolipas 54 Helsinki Suomalaisen

Kirjallisuuden Seura

Petraeligus AEligschillus 1649 Linguaelig Finnicaelig brevis institutio Aboaelig Petrus Wald

Plank Frans 2015 Time for change ndash Carlotta Viti (ed) Perspectives on Historical Syntax

Amsterdam ndash Philadelphia John Benjamins 61ndash91

Qvigstad J K 1881 Beitraumlge zur Vergleichung des verwandten Wortvorrathes der

lappischen und der finnischen Sprache ndash Acta Societatis Scientiarum Fennicaelig 12 113ndash

240

Rask Rasmus 1832 Raeligsonneret lappisk Sproglaeligre efter den Sprogart som bruges af

Fjaeligldlapperne i Porsangerfjorden i Finmarken En Omarbejdelse af Prof Knud Leems

Lappiske grammatica Koslashbenhavn J H Schubothes Boghandling

Raumltsep Huno 1979 Eesti keele ajalooline morfoloogia II Otildepivahend eesti filoloogia

osakonna uumlliotildepilastele Tartu Tartu Riiklik Uumllikool

Ravila Paavo 1935 Die stellung des lappischen innerhalb der finnisch-ugrischen

sprachfamilie ndash Finnisch-Ugrische Forschungen 23 20ndash65

ndashndashndash 1958 Die Ursprache als Grundbegriff der Sprachgeschichte ndash Suomalais-Ugrilaisen


Reacutedei (Radanovics) Kaacuteroly 1962 Die Postpositionen im Syrjaumlnischen unter

Beruumlcksichtigung des Wotjakischen Budapest Akadeacutemiai Kiadoacute


156

Reacutedei Kaacuteroly 1996 Zu der Geschichte des PU-PFU Kasussystems Die Rolle der Koaffixe in

der Herausbildung der Deklination ndash Lars-Gunnar Larsson (ed) Lapponica et uralica

100 Jahre finnisch-ugrischer Unterricht an der Universitaumlt Uppsala Vortraumlge am

Jubilaumlumssymposium 20ndash23 April 1994 Studia Uralica Upsaliensia 26 Uppsala 257ndash

271

Renvall Gustaf 1840 Finsk Spraringklaumlra Enligt den rena Vest-Finska i Bokspraringk vanliga

Dialecten Aringbo Christ Ludv Hjelt

Saarinen Sirkka 2005 Mordwinisch langolanga ndash Finnisch-Ugrische Mitteilungen 2829

321ndash329

Salminen Tapani 1998 Nenets ndash Daniel Abondolo (ed) The Uralic languages London ndash

New York Routledge 516ndash547

ndashndashndash 2002 Problems in the taxonomy of the Uralic languages in the light of modern

comparative studies ndash Лингвистический бепредел сборник статей к 70-летию А

И Кузнецовой Москва Издательство Московского университета 45ndash55

ndashndashndash 2014 Suomalais-samojedilaisia muotovertailuja ndash Nobufumi Inaba Jorma Luutonen

Arja Hamari amp Elina Ahola (eds) Juuret marin murteissa latvus yltaumlauml Uraliin

Juhlakirja Sirkka Saarisen 60-vuotispaumlivaumlksi 21122014 Suomalais-Ugrilaisen Seuran

Toimituksia 270 Helsinki Suomalais-Ugrilainen Seura 289ndash300

Sammallahti Pekka 1977 Norjansaamen Itauml-Enontekioumln murteen aumlaumlnneoppi Suomalais-

Ugrilaisen Seuran Toimituksia 160 Helsinki Suomalais-Ugrilainen Seura

ndashndashndash 1988 Historical Phonology of the Uralic Languages ndash Denis Sinor (ed) The Uralic

Languages Description History and Foreign Influences Leiden ndash New York ndash

Koslashbenhavn ndash Koumlln E J Brill 478ndash554

ndashndashndash 1998 The Saami Languages An Introduction Kaacuteraacutešjohka Davvi Girji

Serebrennikov = Серебренников Б А 1962 Из истории падежной системы пермских

языков ndash В И Лыткин К Е Майтинская amp Б А Серебренников (eds) Вопросы

финно-угорского языкознания К 70-летию со дня рождения члена-

корреспондента АН СССР Д В Бубриха Москва ndash Ленинград Издательство

Академии Наук СССР 9ndash32

ndashndashndash = Серебренников Б А 1963 Историческая морфология пермских языков Москва

Издательство Академии Наук СССР

Setaumllauml E N 1890 Yhteissuomalaisten klusiilien historia Luku yhteissuomalaisesta

aumlaumlnnehistoriasta Helsinki Suomalaisen Kirjallisuuden Seura

Sjoumlgren And Joh 1828 Anteckningar om foumlrsamlingarne i Kemi-Lappmark Helsingfors


157

Sjoumlgren Joh Andreas amp Wiedemann Ferdinand Joh 1861 = Joh Andreas Sjoumlgrenrsquos Livische

Grammatik nebst Sprachproben Im Auftrage der Kaiserlichen Akademie der

Wissenschaften bearbeitet und mit einer historisch-etnographishen Einleitung versehen

von Ferdinand Joh Wiedemann Joh Andreas Sjoumlgrenrsquos gesammelte Schriften Band 2

Theil 1 St Petersburg Kaiserliche Akademie der Wissenschaften

SSA = Itkonen Erkki amp Ulla-Maija Kulonen (eds) 1992ndash2000 Suomen sanojen alkuperauml

Etymologinen sanakirja Suomalaisen Kirjallisuuden Seuran toimituksia 556

Kotimaisten kielten tutkimuskeskuksen julkaisuja 62 Helsinki Kotimaisten kielten

tutkimuskeskus ndash Suomalaisen Kirjallisuuden Seura

Stockfleth N V 1840 Grammatik i det lappiske Sprog saaledes som det tales i Norsk-

Finmarken Foumlrste Del Bogstav- og Formlaeligren Christiania Chr Groumlndahl

Suoniemi-Taipale Inga 1994 Itaumlmerensuomalaisten kielten prolatiivi Suomalaisen

Kirjallisuuden Seuran toimituksia 616 Helsinki Suomalaisen Kirjallisuuden Seura

Szinnyei Josef 1910 Finnisch-ugrische Sprachwissenschaft Leipzig Goumlschenrsquosche

Verlagshandlung

Tauli Valter 1952 Bemerkungen zum Ursprung der uralischen Kasussysteme ndash Ural-

Altaische Jahrbuumlcher 24 3ndash4 27ndash41

ndashndashndash 1956 The origin of affixes ndash Finnisch-Ugrische Forschungen 32 170ndash225

Tereščenko = Терещенко Н М 1965 Ненецко-русский словарь Москва Советская

Энциклопедия

Thordarson Fridrik 2009 Ossetic grammatical studies Edited by Sonja Fritz Wien Verlag

der Oumlsterreichischen Akademie der Wissenschaften

Tikka Toivo 1992 Vepsaumln suffiksoituneet postpositiot Kieliopillisiin sijoihin liittyvauml

suffiksoituminen Studia Uralica Upsaliensia 22 Uppsala

UEW = Reacutedei Kaacuteroly (unter Mitarbeit von Marianne Bakroacute-Nagy Saacutendor Csuacutecs Istvaacuten

Erdeacutelyi Laacuteszloacute Honti Eacuteva Korenchy Eacuteva K Sal und Edit Veacutertes) 1988ndash1991

Uralisches etymologisches Woumlrterbuch 1ndash3 Wiesbaden Otto Harrassowitz

Uotila T E 1945 Sijapaumlaumltteiden syntyhistoriaa ndash Virittaumljauml 49 327ndash336 499ndash501

Vainik Ene 1995 Eesti keele vaumlliskohakaumlaumlnete semantika kognitiivse grammatika

vaatenurgast Tallinn Eesti Teaduste Akadeemia Eesti Keele Instituut

Wichmann Yrjouml 1913ndash1918 Beitraumlge zur tscheremissischen Nominalbildungslehre ndash

Suomalais-Ugrilaisen Seuran Aikakauskirja 306 1ndash42


158

Virtaranta Pertti 1962 Uumlber die Partikeln auf -li in den ostseefi Sprachen ndash Commentationes

Fenno-Ugricae in honorem Paavo Ravila Suomalais-Ugrilaisen Seuran Toimituksia

125 Helsinki 625ndash669

Ylikoski Jussi 2005 Uusia naumlkoumlkulmia suomen infiniittisiin rakenteisiin ndash Virittaumljauml 109

611ndash622

ndashndashndash 2006 Fuomaacutešumit saacutemegiela adposišuvnnaid funkšuvnnain ovdamearkan alde- ja ala-

postposišuvnnaid ii-lokaacutela geavaheapmi ndash Saacutemi dieđalaš aacuteigečaacutela 12006 39ndash61

ndashndashndash 2011 A survey of the origins of directional case suffixes in European Uralic ndash Seppo

Kittilauml Katja Vaumlsti amp Jussi Ylikoski (eds) Case animacy and semantic roles


ndashndashndash 2016 The origins of the western Uralic s-cases revisited historiographical functional-

typological and Samoyedic perspectives ndash Finnisch-Ugrische Forschungen 63 6ndash78

Contact information

Ante Aikio

Saacutemi University of Applied Sciences

Haacutennoluohkkaacute 45

NO-9520 GUOVDAGEAIDNU

Email anteaikiosamiskhsno

Jussi Ylikoski

Department of Language and Culture


PO Box 6050 Langnes

NO-9037 TROMSOslash

Email jussiylikoskiuitno


60

1 Introduction

















1 Introduction

In linguistic literature the Uralic languages are well-known for their large case inventories

Extensive case systems consisting of over ten cases are found in Finnic Mordvin and Permic

languages and in Hungarian Even though such case systems are characteristic of many

modern Uralic languages they are not considered primary to the language family only six

cases are traditionally reconstructed to Proto-Uralic (Janhunen 1982 30ndash31) which is not a

typologically unusual number Hence the question of how the extensive case systems

characteristic of many branches of the family have developed has become a central research

problem in Uralic historical morphology

In the western part of the language family the case system evidently became enriched

already at an early period Through a comparison of Saami Finnic and Mordvin languages

one can reconstruct as many as thirteen cases or case-like suffixes which are reflected in at

least two of these three language branches (see Table 1) The most important innovation

common to these languages (and to Mari as well) involves a reorganization of the local case


61













Finnic

Case Suffix Saami

languages

Finnic

languages

Mordvin

languages


genitive -n + + +

accusative -m + + +




lative -ŋ ( ~ -k -n) (+) (+) +


inessive -snA + + +

elative -stA + + +











62




Table 2




























63









scheme)4














[Finsk]




[ndash ndash]




64






lsquo[Finnish]




[ndash ndash]














Finsk Lappisk








65






Finnish Saami




























66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash



61













Finnic

Case Suffix Saami

languages

Finnic

languages

Mordvin

languages


genitive -n + + +

accusative -m + + +




lative -ŋ ( ~ -k -n) (+) (+) +


inessive -snA + + +

elative -stA + + +











62




Table 2




























63









scheme)4














[Finsk]




[ndash ndash]




64






lsquo[Finnish]




[ndash ndash]














Finsk Lappisk








65






Finnish Saami




























66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash



62




Table 2




























63









scheme)4














[Finsk]




[ndash ndash]




64






lsquo[Finnish]




[ndash ndash]














Finsk Lappisk








65






Finnish Saami




























66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash



63









scheme)4














[Finsk]




[ndash ndash]




64






lsquo[Finnish]




[ndash ndash]














Finsk Lappisk








65






Finnish Saami




























66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash



64






lsquo[Finnish]




[ndash ndash]














Finsk Lappisk








65






Finnish Saami




























66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash



65






Finnish Saami




























66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash



66






















(1869 44)













67












their origin




(Donner 1879 91)




ndash]rsquo






1879 92)








68

































69


































70















260)8











emphasis added)







Section 42 below


71






casesrsquo




























72

































73







put by Alhoniemi



























74


























75

Proto-

Uralic

Tundra

Nenets9

Komi Udmurt Inari

Saami

North

Saami

Lule

Saami

Finnish































transcription


76
































77


















walk1SG toePLADE1SG







stepPST1PL iceALL





78















casesrsquo















79

























(2006)









80































81






1978b 100)





















1992 83)




82






(MAacute 1995)









case form














83



juvllaid nala

wheelPLGA ala






1995)


















2005 27)


84




brookDIMGA ala




























85

paumlaumlllauml


1979 175)






























86


Ylikoski (2006)




largeATTR cityLOC






1990 46)










1990 28)



maacutenas

childish



lapsellinenrsquo16



87

childish










happenPROG












postpositions




88












22ndash27)











triprsquo)




unclear





89


Finnish



b) posture 68 (94) 4 (6)






domains














90






48 76 (= 632)


Lentonoidan poika)

27 47 (= 574)

Total 296 403 (= 734)



















sylissauml

lapINE




91

































92







from each other


























93

(Luke 214)






Lule Saami (AringT)



ednamin

earthINE

Udmurt (VS)



Finnish (Raamattu)





Olonetsian (US)





Estonian (Piibel)





Livonian (UT)




peopleDAT good mood


94

































95








riverrsquo (2) etc

















(Matthew 1618)








96











soje (VS)

thatACC






gatePL winCNG


















97





(Mark 825)

















kutskem (VS)

beginPST23SG












98













(Acts 760)18




voiceCOM






shoutCVB sayPST23SG











99

voiceCOM



[ndash ndash] (UT)





























100
































101




































102



















form talja-lnauml
















103


































104


























the l-cases






105


































106






























107






























108



dialects

(1 Kings 1130)






(Biblia 1642)

(1 Samuel 175)




armor yllauml3SG














alitsialatsi ndash




109











my coat offrsquo


















1999)


110



darr darr

Isaga isa ka























111



















darr darr











112


Table 13















goIMP2SG ovenALL













113




goINF fieldPLALL


























114



usage













Finnic










115




darr darr



darr


cf South Saami

























116



































117





darr


darr


cf South Saami





Proto-Uralic



darr



darr

wolka-n uumll-nauml


darr

tuli-n uumll-nauml


darr



~ jieŋrsquoal

gieđa alde

~ gieđrsquoal

oalggi alde

~ oalggrsquoal

dola alde

~ dolrsquoal

cf (8) (11) cf (1) cf (15) (44) cf (3) (19)








118























(Luke 518)



alde (OT)

alde




119


manACC carryCVB






(Mark 655)
























120





























toe veľďe goINF


goINF toePLADE3SG



121






(56) meľ veľďe

mind velde


mindADE3SG






















122










sonze veľďe)













be aloń instead24







lsquoeggrsquo


123


book be3SG tableADE









1SGADE be3SG book


b Ota kirja minulta



















124





orohagaid ala








(Acts 126)



Mattiasa ala (OT)25

MatthiasGA ala




MattiasALL









125





NOM SG ieš

GEN 1SG iežan

2SG iežat

3SG iežas

LOC 1SG alddaacuten

2SG alddaacutet

3SG alddis

ILL 1SG alccen

2SG alccet

3SG alcces














sidjiide addaacute

3PLILL give3SG




126




1991 19)

(John 7 17)

























127


































128























83)












129


















and Yukaghir26
















case


130




























131


































132



































133




































134















nopeasti )

fastADV

















135








each otherrsquo



otherrsquo



lsquooverrsquo























136





postpositions




























137

































138
































139
































140




































141



































142


























Section 34 above










143




























Telkiaumln pihatse






144


















(MAacute 1995)














145
















Abbreviations

1 first person

2 second person

3 third person

ABE abessive

ABL ablative

ACC accusative

ADE adessive

ADJ adjective

ADV adverb

ALL allative

AN action nominal


ATTR attributive

CAUS causative

CMPV comparative

CNG connegative

COM comitative

COMP complement

CVB converb

DAT dative

DEF definite

DIM diminutive


DU dual

ELA elative

ESS essive

EX existential

FUT future


146


GEN genitive

ILL illative

IMP imperative

INCH inchoative

INE inessive

INF infinitive

INFR inferential

INS instrumental

LAT lative

LOC locative

NEG negative verb

NOM nominative

PASS passive

PL plural


PROG progressive

PROL prolative

PRS present tense

PST past tense


PTCP participle

PTV partitive

Q question

REFL reflexive

SG singular

STEM word stem

SUP superlative

TRANSL translative

References

Corpus material













147






















Davvi Girji os










148











Ougrienne








1990












149

Other references


60 Helsinki


















81ndash87





Ugrilainen Seura








150





Aschehoug amp Co























Cappelen








151


Turun yliopisto












Otava













37 55ndash79








152











77 219ndash274







Series 14 31ndash40





Instituut







tutkimuskeskus






153



































154










287











tutkimuskeskus










Gruyter




155


languages MSFOu 123





















240













156





271




321ndash329



























157















Verlagshandlung



















158





611ndash622








Contact information

Ante Aikio





Jussi Ylikoski



PO Box 6050 Langnes

NO-9037 TROMSOslash


Documents

Ante Aikio & Jussi Ylikoski Sámi University of Applied ... · languages and in Hungarian. Even though such case systems are characteristic of many modern Uralic languages, they are